performance tuning - Page 498

How to tell whether your programmers are under-performing?

- by A Team Lead

I am a team lead with 5+ developers. I have a developer (let's call him A) who is a good programmer, who writes good clean, easy to understand code. However he is somewhat difficult to manage, and sometimes I wonder whether he is really under-performing or not. Our company requires the developers to indicate the work progress in the bug tracker we use, not so much as to monitor the programmers but to let the stackholders know the progress. The thing is, A only updates a task progress when it is done ( maybe 3 weeks after it is first worked on) and this leaves everyone wondering what is going on in the middle of the development week. He wouldn't change his habit despite repeated probing. ( It's OK, developers hate paperwork, I do, too) Recent 2-3 months he on leave quite often due to various events-- either he is sick, or have to attend a lot of personal events etc. ( It's OK, bad things happen in a string. It's just a coincidence) We define sprints, or roadmaps for each month. And in the beginning of the sprint, we will discuss the amount of work each of the developers have to do in a sprint and the developers get to set the amount of time they need for each task. He usually won't be able to complete all of them. (It's OK, the developers are regularly missing deadlines not due to their fault). If only one or two of the above events happen, I won't feel that A is under-performing, but they all happen together. So I have the feeling that A is under-performing and maybe-- God forbid--- slacking off. This is just a feeling based on my years of experience as programmer. But I could be wrong. It is notoriously hard to measure the work of a programmer, given that not all two tasks are alike, and there lacks a standard objective to measure the commitment of a programmer to your company. It is downright impossible to tell whether the programmer is doing his job or slacking off. All you can do, is to trust them-- yeah, trusting and giving them autonomy is the best way for programmers to work, I know that, so don't start a lecture on why you need to trust your programmers, thank you every much-- but if they abuse your trust, can you know? My question is, how can you tell whether your programmers are under-performing? Surely there are experience team leads who know better than me on this? Outcome: I've a straight talk with him regarding my perception on his performance. He was indignant when I suggested that I had the feeling that he wasn't performing at his best level. He felt that this was a completely unfair feeling. I then replied that this was my feeling and I didn't know whether my feeling was right or not. He would have none of this and ended the discussion immediately. Before he left he said that he "would try to give more to the company" in a very cold tone. I was taken aback by his reaction. I am sure that I offended him in some ways. Not too sure whether that was the right thing to do for me to be so frank with him, though. Extra notes: I hate micromanaging. So all that we have for our software process is Sprint ( where tasks get prioritized and assigned, and at the end of the month, a review of the amount of work done). Developers would require to update the tasks as they go along everyday. There is no standup meeting, or anything of the sort. Mainly because we have the freedom to work from home and everyone cherishes this freedom. Although I am the one who sets the deadline, but the developers will provide the estimate for each tasks and I will decide-- based on the estimate-- the tasks that go into a particular sprint. If they can't finish the tasks at the end of the sprint, I will push them to the next. So theoretically one can just do only 1 or 2 tasks during the whole sprint and then push the remaining 99 tasks to the next sprint and still he will be fine as long as justifies this-- in the form of daily work progress updates

Read the article

Windows in StreamInsight: Hopping vs. Snapshot

- by Roman Schindlauer

Three weeks ago, we explained the basic concept of windows in StreamInsight: defining sets of events that serve as arguments for set-based operations, like aggregations. Today, we want to discuss the so-called Hopping Windows and compare them with Snapshot Windows. We will compare these two, because they can serve similar purposes with different behaviors; we will discuss the remaining window type, Count Windows, another time. Hopping (and its syntactic-sugar-sister Tumbling) windows are probably the most straightforward windowing concept in StreamInsight. A hopping window is defined by its length, and the offset from one window to the next. They are aligned with some absolute point on the timeline (which can also be given as a parameter to the window) and create sets of events. The diagram below shows an example of a hopping window with length of 1h and hop size (the offset) of 15 minutes, hence creating overlapping windows: Two aspects in this diagram are important: Since this window is overlapping, an event can fall into more than one windows. If an (interval) event spans a window boundary, its lifetime will be clipped to the window, before it is passed to the set-based operation. That’s the default and currently only available window input policy. (This should only concern you if you are using a time-sensitive user-defined aggregate or operator.) The set-based operation will be applied to each of these sets, yielding a result. This result is: A single scalar value in case of built-in or user-defined aggregates. A subset of the input payloads, in case of the TopK operator. Arbitrary events, when using a user-defined operator. The timestamps of the result are almost always the ones of the windows. Only the user-defined operator can create new events with timestamps. (However, even these event lifetimes are subject to the window’s output policy, which is currently always to clip to the window end.) Let’s assume we were calculating the sum over some payload field: var result = from window in source.HoppingWindow( TimeSpan.FromHours(1), TimeSpan.FromMinutes(15), HoppingWindowOutputPolicy.ClipToWindowEnd) select new { avg = window.Avg(e => e.Value) }; Now each window is reflected by one result event: As you can see, the window definition defines the output frequency. No matter how many or few events we got from the input, this hopping window will produce one result every 15 minutes – except for those windows that do not contain any events at all, because StreamInsight window operations are empty-preserving (more about that another time). The “forced” output for every window can become a performance issue if you have a real-time query with many events in a wide group & apply – let me explain: imagine you have a lot of events that you group by and then aggregate within each group – classical streaming pattern. The hopping window produces a result in each group at exactly the same point in time for all groups, since the window boundaries are aligned with the timeline, not with the event timestamps. This means that the query output will become very bursty, delivering the results of all the groups at the same point in time. This becomes especially obvious if the events are long-lasting, spanning multiple windows each, so that the produced result events do not change their value very often. In such a case, a snapshot window can remedy. Snapshot windows are more difficult to explain than hopping windows: they represent those periods in time, when no event changes occur. In other words, if you mark all event start and and times on your timeline, then you are looking at all snapshot window boundaries: If your events are never overlapping, the snapshot window will not make much sense. It is commonly used together with timestamp modification, which make it a very powerful tool. Or as Allan Mitchell expressed in in a recent tweet: “I used to look at SnapshotWindow() with disdain. Now she is my mistress, the one I turn to in times of trouble and need”. Let’s look at a simple example: I want to compute the average of some value in my events over the last minute. I don’t want this output be produced at fixed intervals, but at soon as it changes (that’s the true event-driven spirit!). The snapshot window will include all currently active event at each point in time, hence we need to extend our original events’ lifetimes into the future: Applying the Snapshot window on these events, it will appear to be “looking back into the past”: If you look at the result produced in this diagram, you can easily prove that, at each point in time, the current event value represents the average of all original input event within the last minute. Here is the LINQ representation of that query, applying the lifetime extension before the snapshot window: var result = from window in source .AlterEventDuration(e => TimeSpan.FromMinutes(1)) .SnapshotWindow(SnapshotWindowOutputPolicy.Clip) select new { avg = window.Avg(e => e.Value) }; With more complex modifications of the event lifetimes you can achieve many more query patterns. For instance “running totals” by keeping the event start times, but snapping their end times to some fixed time, like the end of the day. Each snapshot then “sees” all events that have happened in the respective time period so far. Regards, The StreamInsight Team

Read the article

Session Report - Modern Software Development Anti-Patterns

- by Janice J. Heiss

In this standing-room-only session, building upon his 2011 JavaOne Rock Star “Diabolical Developer” session, Martijn Verburg, this time along with Ben Evans, identified and explored common “anti-patterns” – ways of doing things that keep developers from doing their best work. They emphasized the importance of social interaction and team communication, along with identifying certain psychological pitfalls that lead developers astray. Their emphasis was less on technical coding errors and more how to function well and to keep one’s focus on what really matters. They are the authors of the highly regarded The Well-Grounded Java Developer and are both movers and shakers in the London JUG community and on the Java Community Process. The large room was packed as they gave a fast-moving, witty presentation with lots of laughs and personal anecdotes. Below are a few of the anti-patterns they discussed.Anti-Pattern One: Conference-Driven DeliveryThe theme here is the belief that “Real pros hack code and write their slides minutes before their talks.” Their response to this anti-pattern is an expression popular in the military – PPPPPP, which stands for, “Proper preparation prevents piss-poor performance.”“Communication is very important – probably more important than the code you write,” claimed Verburg. “The more you speak in front of large groups of people the easier it gets, but it’s always important to do dry runs, to present to smaller groups. And important to be members of user groups where you can give presentations. It’s a great place to practice speaking skills; to gain new skills; get new contacts, to network.”They encouraged attendees to record themselves and listen to themselves giving a presentation. They advised them to start with a spouse or friends if need be. Learning to communicate to a group, they argued, is essential to being a successful developer. The emphasis here is that software development is a team activity and good, clear, accessible communication is essential to the functioning of software teams. Anti-Pattern Two: Mortgage-Driven Development The main theme here was that, in a period of worldwide recession and economic stagnation, people are concerned about keeping their jobs. So there is a tendency for developers to treat knowledge as power and not share what they know about their systems with their colleagues, so when it comes time to fix a problem in production, they will be the only one who knows how to fix it – and will have made themselves an indispensable cog in a machine so you cannot be fired. So developers avoid documentation at all costs, or if documentation is required, put it on a USB chip and lock it in a lock box. As in the first anti-pattern, the idea here is that communicating well with your colleagues is essential and documentation is a key part of this. Social interactions are essential. Both Verburg and Evans insisted that increasingly, year by year, successful software development is more about communication than the technical aspects of the craft. Developers who understand this are the ones who will have the most success. Anti-Pattern Three: Distracted by Shiny – Always Use the Latest Technology to Stay AheadThe temptation here is to pick out some obscure framework, try a bit of Scala, HTML5, and Clojure, and always use the latest technology and upgrade to the latest point release of everything. Don’t worry if something works poorly because you are ahead of the curve. Verburg and Evans insisted that there need to be sound reasons for everything a developer does. Developers should not bring in something simply because for some reason they just feel like it or because it’s new. They recommended a site run by a developer named Matt Raible with excellent comparison spread sheets regarding Web frameworks and other apps. They praised it as a useful tool to help developers in their decision-making processes. They pointed out that good developers sometimes make bad choices out of boredom, to add shiny things to their CV, out of frustration with existing processes, or just from a lack of understanding. They pointed out that some code may stay in a business system for 15 or 20 years, but not all code is created equal and some may change after 3 or 6 months. Developers need to know where the code they are contributing fits in. What is its likely lifespan? Anti-Pattern Four: Design-Driven Design The anti-pattern: If you want to impress your colleagues and bosses, use design patents left, right, and center – MVC, Session Facades, SOA, etc. Or the UML modeling suite from IBM, back in the day… Generate super fast code. And the more jargon you can talk when in the vicinity of the manager the better.Verburg shared a true story about a time when he was interviewing a guy for a job and asked him what his previous work was. The interviewee said that he essentially took patterns and uses an approved book of Enterprise Architecture Patterns and applied them. Verburg was dumbstruck that someone could have a job in which they took patterns from a book and applied them. He pointed out that the idea that design is a separate activity is simply wrong. He repeated a saying that he uses, “You should pay your junior developers for the lines of code they write and the things they add; you should pay your senior developers for what they take away.”He explained that by encouraging people to take things away, the code base gets simpler and reflects the actual business use cases developers are trying to solve, as opposed to the framework that is being imposed. He told another true story about a project to decommission a very long system. 98% of the code was decommissioned and people got a nice bonus. But the 2% remained on the mainframe so the 98% reduction in code resulted in zero reduction in costs, because the entire mainframe was needed to run the 2% that was left. There is an incentive to get rid of source code and subsystems when they are no longer needed. The session continued with several more anti-patterns that were equally insightful.

Read the article

SSIS Reporting Pack v0.4 – Execution Report updated

- by jamiet

SSIS Reporting Pack is a suite of reports that I maintain at http://ssisreportingpack.codeplex.com/ that provide visualisation over the SSIS Catalog in SQL Server 2012 and attempt to add value over the reports that ship in the box. Work on the reports has stalled (my last SSIS Reporting Pack blog post was on 4th September 2011) as I’ve had rather more important things going on my life of late however I have recently checked-in a fix that couldn’t really be delayed. I discovered a problem with the Execution report that was causing the report to effectively hang, it was caused by this bit of SQL hidden away in the report definition: [generated_executables] AS ( SELECT [new_executable].[execution_path],[new_executable].[parent_execution_path] FROM ( SELECT [execution_path] = SUBSTRING([loop_iteration].[execution_path] ,1, [loop_iteration].length_exec_path - [loop_iteration].[char_index_close_square] + 1) , [parent_execution_path] = SUBSTRING([loop_iteration].[execution_path] ,1, [loop_iteration].length_exec_path - [loop_iteration].[char_index_open_square]) FROM ( SELECT [execution_path] , [char_index_open_square] = CHARINDEX('[',REVERSE([execution_path]),1) , [char_index_close_square] = CHARINDEX(']',REVERSE([execution_path]),1) , [length_exec_path] = LEN([execution_path]) FROM [exec_stats] es WHERE execution_path LIKE '%\[%]%' ESCAPE '\' )AS [loop_iteration] ) AS [new_executable] GROUP BY [new_executable].[execution_path],[new_executable].[parent_execution_path]) It was there because SSIS does not currently treat a loop iteration as an executable yet I figured there was still value in being able to view it as such – this SQL essentially “invents” new executables for those loop iterations; its what enabled the following visualisation: where each of the three iterations of a For Each Loop called “FEL Loop over top performing regions” appear in the report. Unfortunately, as I alluded, this could under certain circumstances (most likely when there were many loop iterations) cause the report to hang as it waited for the results to be constructed and returned. The change that I have made eradicates this generation of “fake” executables and thus produces this visualisation instead: Notice that the three “children” of the For Each Loop are no longer the three iterations but actually the task (“EPT Call Data Export Package”) contained within that For Each Loop. The problem here is of course that there is no longer a visual distinction between those three iterations; I have instead made the full execution path viewable via a tooltip: If you preferred the “old” way of presenting this information and are happy to put up with the performance degradation then I have kept the old version of the report hanging around in the reporting pack as “execution loop with iterations” however none of the other reports link to it so you will have to browse to it manually if you want to use it. Please let me know if you ARE using it – I would be very interested to hear about your experiences. The last change to make you aware of in the execution report is that by default I no longer show OnPreValidate or OnPostValidate messages as I consider them to be superfluous and only serve to clutter up the results. If you want to put them back, well, its open source so go right ahead! The latest release of SSIS Reporting Pack that contains all of these changes is v0.4 and can be downloaded from http://ssisreportingpack.codeplex.com/releases/view/88178 Feedback on all of the above changes would be very much appreciated. @Jamiet

Read the article

How-to populate different select list content per table row

- by frank.nimphius

Normal 0 false false false EN-US X-NONE X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:"Times New Roman"; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi;} A frequent requirement posted on the OTN forum is to render cells of a table column using instances of af:selectOneChoices with each af:selectOneChoice instance showing different list values. To implement this use case, the select list of the table column is populated dynamically from a managed bean for each row. The table's current rendered row object is accessible in the managed bean using the #{row} expression, where "row" is the value added to the table's var property. <af:table var="row"> ... <af:column ...> <af:selectOneChoice ...> <f:selectItems value="#{browseBean.items}"/> </af:selectOneChoice> </af:column </af:table> The browseBean managed bean referenced in the code snippet above has a setItems and getItems method defined that is accessible from EL using the #{browseBean.items} expression. When the table renders, then the var property variable - the #{row} reference - is filled with the data object displayed in the current rendered table row. The managed bean getItems method returns a List<SelectItem>, which is the model format expected by the f:selectItems tag to populate the af:selectOneChoice list. public void setItems(ArrayList<SelectItem> items) {} //this method is executed for each table row public ArrayList<SelectItem> getItems() { FacesContext fctx = FacesContext.getCurrentInstance(); ELContext elctx = fctx.getELContext(); ExpressionFactory efactory = fctx.getApplication().getExpressionFactory(); ValueExpression ve = efactory.createValueExpression(elctx, "#{row}", Object.class); Row rw = (Row) ve.getValue(elctx); //use one of the row attributes to determine which list to query and //show in the current af:selectOneChoice list // ... ArrayList<SelectItem> alsi = new ArrayList<SelectItem>(); for( ... ){ SelectItem item = new SelectItem(); item.setLabel(...); item.setValue(...); alsi.add(item); } return alsi;} For better performance, the ADF Faces table stamps it data rows. Stamping means that the cell renderer component - af:selectOneChoice in this example - is instantiated once for the column and then repeatedly used to display the cell data for individual table rows. This however means that you cannot refresh a single select one choice component in a table to change its list values. Instead the whole table needs to be refreshed, rerunning the managed bean list query. Be aware that having individual list values per table row is an expensive operation that should be used only on small tables for Business Services with low latency data fetching (e.g. ADF Business Components and EJB) and with server side caching strategies for the queried data (e.g. storing queried list data in a managed bean in session scope).

Read the article

yield – Just yet another sexy c# keyword?

- by George Mamaladze

yield (see NSDN c# reference) operator came I guess with .NET 2.0 and I my feeling is that it’s not as wide used as it could (or should) be. I am not going to talk here about necessarity and advantages of using iterator pattern when accessing custom sequences (just google it). Let’s look at it from the clean code point of view. Let's see if it really helps us to keep our code understandable, reusable and testable. Let’s say we want to iterate a tree and do something with it’s nodes, for instance calculate a sum of their values. So the most elegant way would be to build a recursive method performing a classic depth traversal returning the sum. private int CalculateTreeSum(Node top) { int sumOfChildNodes = 0; foreach (Node childNode in top.ChildNodes) { sumOfChildNodes += CalculateTreeSum(childNode); } return top.Value + sumOfChildNodes; } “Do One Thing” Nevertheless it violates one of the most important rules “Do One Thing”. Our method CalculateTreeSum does two things at the same time. It travels inside the tree and performs some computation – in this case calculates sum. Doing two things in one method is definitely a bad thing because of several reasons: · Understandability: Readability / refactoring · Reuseability: when overriding - no chance to override computation without copying iteration code and vice versa. · Testability: you are not able to test computation without constructing the tree and you are not able to test correctness of tree iteration. I want to spend some more words on this last issue. How do you test the method CalculateTreeSum when it contains two in one: computation & iteration? The only chance is to construct a test tree and assert the result of the method call, in our case the sum against our expectation. And if the test fails you do not know wether was the computation algorithm wrong or was that the iteration? At the end to top it all off I tell you: according to Murphy’s Law the iteration will have a bug as well as the calculation. Both bugs in a combination will cause the sum to be accidentally exactly the same you expect and the test will PASS. J Ok let’s use yield! That’s why it is generally a very good idea not to mix but isolate “things”. Ok let’s use yield! private int CalculateTreeSumClean(Node top) { IEnumerable<Node> treeNodes = GetTreeNodes(top); return CalculateSum(treeNodes); } private int CalculateSum(IEnumerable<Node> nodes) { int sumOfNodes = 0; foreach (Node node in nodes) { sumOfNodes += node.Value; } return sumOfNodes; } private IEnumerable<Node> GetTreeNodes(Node top) { yield return top; foreach (Node childNode in top.ChildNodes) { foreach (Node currentNode in GetTreeNodes(childNode)) { yield return currentNode; } } } Two methods does not know anything about each other. One contains calculation logic another jut the iteration logic. You can relpace the tree iteration algorithm from depth traversal to breath trevaersal or use stack or visitor pattern instead of recursion. This will not influence your calculation logic. And vice versa you can relace the sum with product or do whatever you want with node values, the calculateion algorithm is not aware of beeng working on some tree or graph. How about not using yield? Now let’s ask the question – what if we do not have yield operator? The brief look at the generated code gives us an answer. The compiler generates a 150 lines long class to implement the iteration logic. [CompilerGenerated] private sealed class <GetTreeNodes>d__0 : IEnumerable<Node>, IEnumerable, IEnumerator<Node>, IEnumerator, IDisposable { ... 150 Lines of generated code ... } Often we compromise code readability, cleanness, testability, etc. – to reduce number of classes, code lines, keystrokes and mouse clicks. This is the human nature - we are lazy. Knowing and using such a sexy construct like yield, allows us to be lazy, write very few lines of code and at the same time stay clean and do one thing in a method. That's why I generally welcome using staff like that. Note: The above used recursive depth traversal algorithm is possibly the compact one but not the best one from the performance and memory utilization point of view. It was taken to emphasize on other primary aspects of this post.

Read the article

XBRL - Moving from Production to Consumption

- by jmorourke

Here's an update on what’s new with XBRL and how it can actually benefit your organization versus adding extra time and costs to financial reporting. On February 29th (leap day) of 2012 I attended the XBRL and Financial Analysis Technology Conference at Baruch College in NYC. The event, which attracted over 300 XBRL gurus and fans was presented by XBRL US, The New York Society of Security Analysts’ Improved Corporate Reporting Committee, and Baruch College’s Robert Zicklin Center for Corporate Integrity. The event featured keynotes from the U.S. Securities and Exchange Commission (SEC), and the CFA Institute as well as panels covering alternative research tools and data, corporate reporting to stakeholders and a demonstration of XBRL analysis tools. The program culminated in a presentation of the finalists and the winner of the $20,000 XBRL Challenge. Some of the key points made in the sessions included: The focus of XBRL tools is moving from production to consumption. As of February 2012, over 9000 companies are reporting in XBRL, with over 10 million facts filed to date XBRL taxonomy extensions have dropped from 27% to 11% making comparisons easier The SEC reports that XBRL makes it easier to analyze disclosures, focus on accounting issues XBRL is helping standards-setters like the FASB speed their analysis of impacts of proposed accounting rule changes Companies like Thomson Reuters report that XBRL is helping speed the delivery of data to clients The most interesting part of the program though, was the session highlighting the 5 finalists in the XBRL Challenge competition and the winning solution. The XBRL Challenge was launched in 2011 as a means of spurring the development of more end-user tools to help with the consumption of XBRL-based financial information. Over an 8-month process handled by 5 judges, there were 84 registrants, 15 completed submissions, 5 finalists and one winner of the challenge. All of the solutions are open-sourced tools and most of them focus on consuming XBRL-based data. The 5 finalists included: Advanced XBRL Processing from Oxide solutions – XBRL viewer for taxonomies, filings and company data with peer comparison capabilities. Arrelle – API for XBRL processes, supports SEC Validations, RSS Feeds to access filings etc. Calcbench – XBRL data analysis tool that can be embedded in other web applications. This tool can combine XBRL filings with real-time market data. XBRL to XL – allows the importing of XBRL data into Microsoft Excel for analysis, comparisons. Users start on the web and populate Excel with XBRL data. XBurble – allows users to search and view XBRL filings, export to Excel, merge for comparison, and includes a workflow interface. The winner of the $20,000 XBRL Challenge prize was CalcBench. More information about the XBRL Challenge and the finalists can be found at www.XBRLUS.org/challenge XBRL for Sustainability Reporting – other recent news on the XBRL front was the announcement by the Global Reporting Initiative (GRI) of an XBRL taxonomy for Sustainability Reporting. This taxonomy was co-developed by the GRI and Deloitte and is designed to make the consumption of data found in Sustainability Reports much easier. Although there is no government mandate to file Sustainability Reports in XBRL format, organizations that do use the GRI guidelines for Sustainability Reporting are encouraged to tag and submit their data voluntarily to the GRI – who will populate a database with Sustainability Reporting data and make this available to the public. For more information about this initiative, you can go to the GRI web site: www.globalreporting.org. So how does all of this benefit corporate filers and investors? Since its introduction, the consensus in the market is that XBRL has mainly benefited the regulators and investment analysts who need to consume and analyze large volumes of financial data. But with the emergence of more end-user tools for consuming and analyzing XBRL-based data, and the ability to perform quick comparisons of one company versus its peers and competitors in an industry group, will soon accelerate the benefits to corporate finance staff, as well as individual investors. This could apply to financial results tagged in XBRL, as well as non-financial information such as Sustainability Reporting – which over the long-term will likely be integrated with financial reporting. And as multiple regulators and agencies in a country adopt the XBRL standard for corporate filings, more benefits will accrue as companies will be able to leverage one set of XBRL-based financial data for multiple regulatory filings. For more information about the latest developments in XBRL, check out the XBRL US or XBRL International web sites: www.xbrl.org, www.xbrlus.org. For more information about what Oracle is doing to support XBRL, here are some links: http://www.oracle.com/us/solutions/ent-performance-bi/disclosure-management-065892.html http://www.oracle.com/technetwork/database/features/xmldb/index-087631.html Feel free to contact me if you have any questions or need more information: [email protected]

Read the article

await, WhenAll, WaitAll, oh my!!

- by cibrax

If you are dealing with asynchronous work in .NET, you might know that the Task class has become the main driver for wrapping asynchronous calls. Although this class was officially introduced in .NET 4.0, the programming model for consuming tasks was much more simplified in C# 5.0 in .NET 4.5 with the addition of the new async/await keywords. In a nutshell, you can use these keywords to make asynchronous calls as if they were sequential, and avoiding in that way any fork or callback in the code. The compiler takes care of the rest. I was yesterday writing some code for making multiple asynchronous calls to backend services in parallel. The code looked as follow, var allResults = new List<Result>(); foreach(var provider in providers) { var results = await provider.GetResults(); allResults.AddRange(results); } return allResults; You see, I was using the await keyword to make multiple calls in parallel. Something I did not consider was the overhead this code implied after being compiled. I started an interesting discussion with some smart folks in twitter. One of them, Tugberk Ugurlu, had the brilliant idea of actually write some code to make a performance comparison with another approach using Task.WhenAll. There are two additional methods you can use to wait for the results of multiple calls in parallel, WhenAll and WaitAll. WhenAll creates a new task and waits for results in that new task, so it does not block the calling thread. WaitAll, on the other hand, blocks the calling thread. This is the code Tugberk initially wrote, and I modified afterwards to also show the results of WaitAll. class Program { private static Func<Stopwatch, Task>[] funcs = new Func<Stopwatch, Task>[] { async (watch) => { watch.Start(); await Task.Delay(1000); Console.WriteLine("1000 one has been completed."); }, async (watch) => { await Task.Delay(1500); Console.WriteLine("1500 one has been completed."); }, async (watch) => { await Task.Delay(2000); Console.WriteLine("2000 one has been completed."); watch.Stop(); Console.WriteLine(watch.ElapsedMilliseconds + "ms has been elapsed."); } }; static void Main(string[] args) { Console.WriteLine("Await in loop work starts..."); DoWorkAsync().ContinueWith(task => { Console.WriteLine("Parallel work starts..."); DoWorkInParallelAsync().ContinueWith(t => { Console.WriteLine("WaitAll work starts..."); WaitForAll(); }); }); Console.ReadLine(); } static async Task DoWorkAsync() { Stopwatch watch = new Stopwatch(); foreach (var func in funcs) { await func(watch); } } static async Task DoWorkInParallelAsync() { Stopwatch watch = new Stopwatch(); await Task.WhenAll(funcs[0](watch), funcs[1](watch), funcs[2](watch)); } static void WaitForAll() { Stopwatch watch = new Stopwatch(); Task.WaitAll(funcs[0](watch), funcs[1](watch), funcs[2](watch)); } } After running this code, the results were very concluding. Await in loop work starts... 1000 one has been completed. 1500 one has been completed. 2000 one has been completed. 4532ms has been elapsed. Parallel work starts... 1000 one has been completed. 1500 one has been completed. 2000 one has been completed. 2007ms has been elapsed. WaitAll work starts... 1000 one has been completed. 1500 one has been completed. 2000 one has been completed. 2009ms has been elapsed. The await keyword in a loop does not really make the calls in parallel.

Read the article

Pace Layering Comes Alive

- by Tanu Sood

v\:* {behavior:url(#default#VML);} o\:* {behavior:url(#default#VML);} w\:* {behavior:url(#default#VML);} .shape {behavior:url(#default#VML);} Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Calibri","sans-serif"; mso-bidi-font-family:"Times New Roman";} Rick Beers is Senior Director of Product Management for Oracle Fusion Middleware. Prior to joining Oracle, Rick held a variety of executive operational positions at Corning, Inc. and Bausch & Lomb. With a professional background that includes senior management positions in manufacturing, supply chain and information technology, Rick brings a unique set of experiences to cover the impact that technology can have on business models, processes and organizations. Rick hosts the IT Leaders Editorial on a monthly basis. By now, readers of this column are quite familiar with Oracle AppAdvantage, a unified framework of middleware technologies, infrastructure and applications utilizing a pace layered approach to enterprise systems platforms. 1. Standardize and Consolidate core Enterprise Applications by removing invasive customizations, costly workarounds and the complexity that multiple instances creates. 2. Move business specific processes and applications to the Differentiate Layer, thus creating greater business agility with process extensions and best of breed applications managed by cross- application process orchestration. 3. The Innovate Layer contains all the business capabilities required for engagement, collaboration and intuitive decision making. This is the layer where innovation will occur, as people engage one another in a secure yet open and informed way. 4. Simplify IT by minimizing complexity, improving performance and lowering cost with secure, reliable and managed systems across the entire Enterprise. But what hasn’t been discussed is the pace layered architecture that Oracle AppAdvantage adopts. What is it, what are its origins and why is it relevant to enterprise scale applications and technologies? It’s actually a fascinating tale that spans the past 20 years and a basic understanding of it provides a wonderful context to what is evolving as the future of enterprise systems platforms. It all begins in 1994 with a book by noted architect Stewart Brand, of ’Whole Earth Catalog’ fame. In his 1994 book How Buildings Learn, Brand popularized the term ‘Shearing Layers’, arguing that any building is actually a hierarchy of pieces, each of which inherently changes at different rates. In 1997 he produced a 6 part BBC Series adapted from the book, in which Part 6 focuses on Shearing Layers. In this segment Brand begins to introduce the concept of ‘pace’. Brand further refined this idea in his subsequent book, The Clock of the Long Now, which began to link the concept of Shearing Layers to computing and introduced the term ‘pace layering’, where he proposes that: “An imperative emerges: an adaptive [system] has to allow slippage between the differently-paced systems … otherwise the slow systems block the flow of the quick ones and the quick ones tear up the slow ones with their constant change. Embedding the systems together may look efficient at first but over time it is the opposite and destructive as well.” In 2000, IBM architects Ian Simmonds and David Ing published a paper entitled A Shearing Layers Approach to Information Systems Development, which applied the concept of Shearing Layers to systems design and development. It argued that at the time systems were still too rigid; that they constrained organizations by their inability to adapt to changes. The findings in the Conclusions section are particularly striking: “Our starting motivation was that enterprises need to become more adaptive, and that an aspect of doing that is having adaptable computer systems. The challenge is then to optimize information systems development for change (high maintenance) rather than stability (low maintenance). Our response is to make it explicit within software engineering the notion of shearing layers, and explore it as the principle that systems should be built to be adaptable in response to the qualitatively different rates of change to which they will be subjected. This allows us to separate functions that should legitimately change relatively slowly and at significant cost from that which should be changeable often, quickly and cheaply.” The problem at the time of course was that this vision of adaptable systems was simply not possible within the confines of 1st generation ERP, which were conceived, designed and developed for standardization and compliance. It wasn’t until the maturity of open, standards based integration, and the middleware innovation that followed, that pace layering became an achievable goal. And Oracle is leading the way. Oracle’s AppAdvantage framework makes pace layering come alive by taking a strategic vision 20 years in the making and transforming it to a reality. It allows enterprises to retain and even optimize their existing ERP systems, while wrapping around those ERP systems three layers of capabilities that inherently adapt as needed, at a pace that’s optimal for the enterprise.

Read the article

Vendors: Partners or Salespeople?

- by BuckWoody

I got a great e-mail from a friend that asked about how he could foster a better relationship with his vendors. So many times when you work with a vendor it’s more of a used-car sales experience than a partnership – but you can actually make your vendor more of a partner, as long as you both set some ground-rules at the start. Sit down with your vendor, and have a heart-to-heart talk with them, explain that they won’t win every time, but that you’re willing to work with them in an honest way on both sides. Here’s the advice I sent him verbatim. I hope this post generates lots of comments from both customers and vendors. I don’t expect that you’ve had a great experience with your Microsoft reps, but I happen to work with some of the best sales teams in the business, and our clients tell us that all the time. “The key to this relationship is to keep the audience really small. Ideally there should be one person from your side that is responsible for the relationship, and one from the vendor’s side. Each responsible person should have the authority to make decisions, and to bring in other folks as needed for a given topic, project or decision. For Microsoft, this is called an “Account Manager” – they aren’t technical, they aren’t sales. They “own” a relationship with a company. They learn what the company does, who does it, and how. They are responsible to understand what the challenges in your company are. While they don’t know the bits and bytes of everything we sell, they know what each thing does, and who to talk to about it. I get a call from an Account Manager every week that has pre-digested an issue at an organization and says to me: “I need you to set up an architectural meeting with their technical staff to get a better read on how we can help with problem X.” I do that and then report back to the Account Manager what we learned. All through this process there’s the atmosphere of a “team”, not a “sales opportunity” per se. I’ve even recommended that the firm use a rival product, and I’ve never gotten push-back on that decision from my Account Managers. But that brings up an interesting point. Someone pays an Account Manager and pays me. They expect something in return. At some point, you have to buy something. Not every time, not every situation – sometimes it’s just helping you with what you already bought from us. But the point is that you can’t expect lots of love and never spend any money. That’s the way business works. Finally, don’t view the vendor as someone with their hand in your pocket – somebody that’s just trying to sell you something and doesn’t care if they ever see you again – unless they deserve it. There are plenty of “love them and leave them” companies out there, and you may have even had this experience with us, but that isn’t the case in the firms I work with. In fact, my customers get a questionnaire that asks them that exact question. “How many times have you seen your account team? Did you like your interaction with them? Can they do better?” My raises, performance reviews and general standing in my group are based on the answers the company gives. Ask your vendor if they measure their sales and support teams this way – if not, seek another vendor to partner with. Partnering with someone is a big deal. It involves time and effort on your part, and on the vendor’s part. If either of you isn’t pulling your weight, it just won’t work. You have every right to expect them to treat you as a partner, and they have the same right for your side.” Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!

Read the article

Drive Online Engagement with Intuitive Portals and Websites

- by kellsey.ruppel

As more and more business is being conducted via online channels, engaging users and making them more productive and efficient though these online channels is becoming critical. These users could be customers, partners or employees and while the respective channels through which they interact might be different, these users do increasingly interact with your business through the Web, or mobile devices or now through various social mediums. Businesses need a user engagement strategy and solution that allows them to deliver targeted and personalized content and applications to users through the various online mediums and touch points. The customer experience today is made up of an ongoing set of interactions with organizations across many channels, online and offline. The Direct channel (including sales reps, email and mail) is an important point of contact, as is the Contact Center. Contact Centers rely on the phone as a means of interacting with customers, and also more now than ever, the Web as well. However, the online organization is often managed separately from the Contact Center organization within a business. In-store is an important channel for retailers, offering Point-of-Service for human interactions, and Kiosks which enable self-service. Kiosks are a Web-enabled touch point but in-store kiosks are often managed by the head of retail operations, rather than the online organization. And of course, the online channel, including customer interactions with an organization via digital means -- on the website, mobile websites, and social networking sites, has risen to paramount importance in recent years in the customer experience. Historically all of these channels have been managed separately. The result of all of this fragmentation is that the customer touch points with an organization are siloed. Their interactions online are not known and respected in their dealings in-store. Their calls to the contact center are not taken as input into what the website offers them when they arrive. Think of how many times you’ve fallen victim to this. Your experience with the company call center is different than the experience in-store. Your experience with the company website on your desktop computer is different than your experience on your iPad. I think you get the point. But the customer isn’t the only one we need to look at here, as employees and the IT organization have challenges as well when it comes to online engagement. There are many common tools and technologies that organizations have been using to try and engage users, whether it’s customers, employees or partners. Some have adopted different blog and wiki technologies (some hosted, some open source, sometimes embedded in platforms), to things like tagging, file sharing and content management, or composite applications for self-service applications and activity streams. Basically, there are so many different tools & technologies that each address different aspects of user engagement. Now, one of the challenges with this, is that if we look at each individual tool, typically just implementing for example a file sharing and basic collaboration solution, may meet the needs of the business user for one aspect of user engagement, but it may not be the best solution to engage with customers and partners, or it may not fit with IT standards such as integrating with their single sign on tools or their corporate website. Often, the scenario is that businesses are having to acquire multiple pieces and parts as well as build custom applications to meet their needs. Leaving customers and partners with a more fragmented way of interacting with the company. Every organization has some sort of enterprise balancing act between the needs of the business user and the needs and restrictions enforced by enterprise IT groups. As we’ve been discussing, we all know that the expectations for online engagement have changed since the days of the static, one-size fits all website. With these changes have come some very difficult organizational challenges as well. Today, as a business user, you want to engage with your customers, and your customers expect you to know who they are. They expect you to recall the details they’ve provided to you on your website, to your CSRs and to your sales people. They expect you to remember their purchases, their preferences and their problems. And they expect you to know who they are, equally well, across channels, including your web presence. This creates a host of challenges for today’s business users. Delivering targeted, relevant content online is now essential for converting prospects into customers and for engendering long term loyalty. Business users need the ability to leverage customer data from different sources to fuel their segmentation and targeting strategies and to easily set-up, manage and optimize online campaigns. Also critical, they need the ability to accomplish these things on-the-fly, at the speed of the marketplace, while making iterative improvements. These changing expectations put a host of demands on the IT organization as well. The web presence must be able to scale to support the delivery of personalized and targeted content to thousands of site visitors without sacrificing performance. And integration between systems becomes more important as well, as organizations strive to obtain one view of the customer culled from WCM data, CRM data and more. So then, how do you solve these challenges and meet the growing demands of your users? You need a solution that: Unifies every customer interaction across all channels Personalizes the products and content that interest the customer and to the device Delivers targeted promotions to the right customer Engages and improve employee productivity Provides self-service access to applications Includes embedded in-context social So how then do you achieve this level of online engagement, complete customer experience and engage your employees? The answer: Oracle WebCenter. If you want to learn how to get there, we encourage you to attend this webcast on Thursday Drive Online Engagement with Intuitive Portals and Websites, where we'll talk about how you are able to transform your portal experience and optimize online engagement -- making your portals more interactive and more engaging across multiple channels. Register today!

Read the article

C#/.NET Little Wonders: Getting Caller Information

- by James Michael Hare

Originally posted on: http://geekswithblogs.net/BlackRabbitCoder/archive/2013/07/25/c.net-little-wonders-getting-caller-information.aspx Once again, in this series of posts I look at the parts of the .NET Framework that may seem trivial, but can help improve your code by making it easier to write and maintain. The index of all my past little wonders posts can be found here. There are times when it is desirable to know who called the method or property you are currently executing. Some applications of this could include logging libraries, or possibly even something more advanced that may server up different objects depending on who called the method. In the past, we mostly relied on the System.Diagnostics namespace and its classes such as StackTrace and StackFrame to see who our caller was, but now in C# 5, we can also get much of this data at compile-time. Determining the caller using the stack One of the ways of doing this is to examine the call stack. The classes that allow you to examine the call stack have been around for a long time and can give you a very deep view of the calling chain all the way back to the beginning for the thread that has called you. You can get caller information by either instantiating the StackTrace class (which will give you the complete stack trace, much like you see when an exception is generated), or by using StackFrame which gets a single frame of the stack trace. Both involve examining the call stack, which is a non-trivial task, so care should be done not to do this in a performance-intensive situation. For our simple example let's say we are going to recreate the wheel and construct our own logging framework. Perhaps we wish to create a simple method Log which will log the string-ified form of an object and some information about the caller. We could easily do this as follows: 1: static void Log(object message) 2: { 3: // frame 1, true for source info 4: StackFrame frame = new StackFrame(1, true); 5: var method = frame.GetMethod(); 6: var fileName = frame.GetFileName(); 7: var lineNumber = frame.GetFileLineNumber(); 8: 9: // we'll just use a simple Console write for now 10: Console.WriteLine("{0}({1}):{2} - {3}", 11: fileName, lineNumber, method.Name, message); 12: } So, what we are doing here is grabbing the 2nd stack frame (the 1st is our current method) using a 2nd argument of true to specify we want source information (if available) and then taking the information from the frame. This works fine, and if we tested it out by calling from a file such as this: 1: // File c:\projects\test\CallerInfo\CallerInfo.cs 2: 3: public class CallerInfo 4: { 5: Log("Hello Logger!"); 6: } We'd see this: 1: c:\projects\test\CallerInfo\CallerInfo.cs(5):Main - Hello Logger! This works well, and in fact CallStack and StackFrame are still the best ways to examine deeper into the call stack. But if you only want to get information on the caller of your method, there is another option… Determining the caller at compile-time In C# 5 (.NET 4.5) they added some attributes that can be supplied to optional parameters on a method to receive caller information. These attributes can only be applied to methods with optional parameters with explicit defaults. Then, as the compiler determines who is calling your method with these attributes, it will fill in the values at compile-time. These are the currently supported attributes available in the System.Runtime.CompilerServices namespace": CallerFilePathAttribute – The path and name of the file that is calling your method. CallerLineNumberAttribute – The line number in the file where your method is being called. CallerMemberName – The member that is calling your method. So let’s take a look at how our Log method would look using these attributes instead: 1: static int Log(object message, 2: [CallerMemberName] string memberName = "", 3: [CallerFilePath] string fileName = "", 4: [CallerLineNumber] int lineNumber = 0) 5: { 6: // we'll just use a simple Console write for now 7: Console.WriteLine("{0}({1}):{2} - {3}", 8: fileName, lineNumber, memberName, message); 9: } Again, calling this from our sample Main would give us the same result: 1: c:\projects\test\CallerInfo\CallerInfo.cs(5):Main - Hello Logger! However, though this seems the same, there are a few key differences. First of all, there are only 3 supported attributes (at this time) that give you the file path, line number, and calling member. Thus, it does not give you as rich of detail as a StackFrame (which can give you the calling type as well and deeper frames, for example). Also, these are supported through optional parameters, which means we could call our new Log method like this: 1: // They're defaults, why not fill 'em in 2: Log("My message.", "Some member", "Some file", -13); In addition, since these attributes require optional parameters, they cannot be used in properties, only in methods. These caveats aside, they do let you get similar information inside of methods at a much greater speed! How much greater? Well lets crank through 1,000,000 iterations of each. instead of logging to console, I’ll return the formatted string length of each. Doing this, we get: 1: Time for 1,000,000 iterations with StackTrace: 5096 ms 2: Time for 1,000,000 iterations with Attributes: 196 ms So you see, using the attributes is much, much faster! Nearly 25x faster in fact. Summary There are a few ways to get caller information for a method. The StackFrame allows you to get a comprehensive set of information spanning the whole call stack, but at a heavier cost. On the other hand, the attributes allow you to quickly get at caller information baked in at compile-time, but to do so you need to create optional parameters in your methods to support it. Technorati Tags: Little Wonders,CSharp,C#,.NET,StackFrame,CallStack,CallerFilePathAttribute,CallerLineNumberAttribute,CallerMemberName

Read the article

Best Practices - Dynamic Reconfiguration

- by jsavit

This post is one of a series of "best practices" notes for Oracle VM Server for SPARC (formerly named Logical Domains) Overview of dynamic Reconfiguration Oracle VM Server for SPARC supports Dynamic Reconfiguration (DR), making it possible to add or remove resources to or from a domain (virtual machine) while it is running. This is extremely useful because resources can be shifted to or from virtual machines in response to load conditions without having to reboot or interrupt running applications. For example, if an application requires more CPU capacity, you can add CPUs to improve performance, and remove them when they are no longer needed. You can use even use Dynamic Resource Management (DRM) policies that automatically add and remove CPUs to domains based on load. How it works (in broad general terms) Dynamic Reconfiguration is done in coordination with Solaris, which recognises a hypervisor request to change its virtual machine configuration and responds appropriately. In essence, Solaris receives a message saying "you now have 16 more CPUs numbered 16 to 31" or "8GB more RAM starting at address X" or "here's a new network or disk device - have fun with it". These actions take very little time. Solaris then can start using the new resource. In the case of added CPUs, that means dispatching processes and potentially binding interrupts to the new CPUs. For memory, Solaris adds the new memory pages to its "free" list and starts using them. Comparable actions occur with network and disk devices: they are recognised by Solaris and then used. Removing is the reverse process: after receiving the DR message to free specific CPUs, Solaris unbinds interrupts assigned to the CPUs and stops dispatching process threads. That takes very little time. primary # ldm list NAME STATE FLAGS CONS VCPU MEMORY UTIL UPTIME primary active -n-cv- SP 16 4G 1.0% 6d 22h 29m ldom1 active -n---- 5000 16 8G 0.9% 6h 59m primary # ldm set-core 5 ldom1 primary # ldm list NAME STATE FLAGS CONS VCPU MEMORY UTIL UPTIME primary active -n-cv- SP 16 4G 0.2% 6d 22h 29m ldom1 active -n---- 5000 40 8G 0.1% 6h 59m primary # ldm set-core 2 ldom1 primary # ldm list NAME STATE FLAGS CONS VCPU MEMORY UTIL UPTIME primary active -n-cv- SP 16 4G 1.0% 6d 22h 29m ldom1 active -n---- 5000 16 8G 0.9% 6h 59m Memory pages are vacated by copying their contents to other memory locations and wiping them clean. Solaris may have to swap memory contents to disk if the remaining RAM isn't enough to hold all the contents. For this reason, deallocating memory can take longer on a loaded system. Even on a lightly loaded system it took several 7 or 8 seconds to switch the domain below between 8GB and 24GB of RAM. primary # ldm set-mem 24g ldom1 primary # ldm list NAME STATE FLAGS CONS VCPU MEMORY UTIL UPTIME primary active -n-cv- SP 16 4G 0.1% 6d 22h 36m ldom1 active -n---- 5000 16 24G 0.2% 7h 6m primary # ldm set-mem 8g ldom1 primary # ldm list NAME STATE FLAGS CONS VCPU MEMORY UTIL UPTIME primary active -n-cv- SP 16 4G 0.7% 6d 22h 37m ldom1 active -n---- 5000 16 8G 0.3% 7h 7m What if the device is in use? (this is the anecdote that inspired this blog post) If CPU or memory is being removed, releasing it pretty straightforward, using the method described above. The resources are released, and Solaris continues with less capacity. It's not as simple with a network or I/O device: you don't want to yank a device out from underneath an application that might be using it. In the following example, I've added a virtual network device to ldom1 and want to take it away, even though it's been plumbed. primary # ldm rm-vnet vnet19 ldom1 Guest LDom returned the following reason for failing the operation: Resource Information ---------------------------------------------------------- ----------------------- /devices/virtual-devices@100/channel-devices@200/network@1 Network interface net1 VIO operation failed because device is being used in LDom ldom1 Failed to remove VNET instance That's what I call a helpful error message - telling me exactly what was wrong. In this case the problem is easily solved. I know this NIC is seen in the guest as net1 so: ldom1 # ifconfig net1 down unplumb Now I can dispose of it, and even the virtual switch I had created for it: primary # ldm rm-vnet vnet19 ldom1 primary # ldm rm-vsw primary-vsw9 If I had to take away the device disruptively, I could have used ldm rm-vnet -f but that could disrupt whoever was using it. It's better if that can be avoided. Summary Oracle VM Server for SPARC provides dynamic reconfiguration, which lets you modify a guest domain's CPU, memory and I/O configuration on the fly without reboot. You can add and remove resources as needed, and even automate this for CPUs by setting up resource policies. Taking things away can be more complicated than giving, especially for devices like disks and networks that may contain application and system state or be involved in a transaction. LDoms and Solaris cooperative work together to coordinate resource allocation and de-allocation in a safe and effective way. For best practices, use dynamic reconfiguration to make the best use of your system's resources.

Read the article

Data management in unexpected places

- by Ashok_Ora

Normal 0 false false false EN-US X-NONE X-NONE Data management in unexpected places When you think of network switches, routers, firewall appliances, etc., it may not be obvious that at the heart of these kinds of solutions is an engine that can manage huge amounts of data at very high throughput with low latencies and high availability. Consider a network router that is processing tens (or hundreds) of thousands of network packets per second. So what really happens inside a router? Packets are streaming in at the rate of tens of thousands per second. Each packet has multiple attributes, for example, a destination, associated SLAs etc. For each packet, the router has to determine the address of the next “hop” to the destination; it has to determine how to prioritize this packet. If it’s a high priority packet, then it has to be sent on its way before lower priority packets. As a consequence of prioritizing high priority packets, lower priority data packets may need to be temporarily stored (held back), but addressed fairly. If there are security or privacy requirements associated with the data packet, those have to be enforced. You probably need to keep track of statistics related to the packets processed (someone’s sure to ask). You have to do all this (and more) while preserving high availability i.e. if one of the processors in the router goes down, you have to have a way to continue processing without interruption (the customer won’t be happy with a “choppy” VoIP conversation, right?). And all this has to be achieved without ANY intervention from a human operator – the router is most likely to be in a remote location – it must JUST CONTINUE TO WORK CORRECTLY, even when bad things happen. How is this implemented? As soon as a packet arrives, it is interpreted by the receiving software. The software decodes the packet headers in order to determine the destination, kind of packet (e.g. voice vs. data), SLAs associated with the “owner” of the packet etc. It looks up the internal database of “rules” of how to process this packet and handles the packet accordingly. The software might choose to hold on to the packet safely for some period of time, if it’s a low priority packet. Ah – this sounds very much like a database problem. For each packet, you have to minimally · Look up the most efficient next “hop” towards the destination. The “most efficient” next hop can change, depending on latency, availability etc. · Look up the SLA and determine the priority of this packet (e.g. voice calls get priority over data ftp) · Look up security information associated with this data packet. It may be necessary to retrieve the context for this network packet since a network packet is a small “slice” of a session. The context for the “header” packet needs to be stored in the router, in order to make this work. · If the priority of the packet is low, then “store” the packet temporarily in the router until it is time to forward the packet to the next hop. · Update various statistics about the packet. In most cases, you have to do all this in the context of a single transaction. For example, you want to look up the forwarding address and perform the “send” in a single transaction so that the forwarding address doesn’t change while you’re sending the packet. So, how do you do all this? Berkeley DB is a proven, reliable, high performance, highly available embeddable database, designed for exactly these kinds of usage scenarios. Berkeley DB is a robust, reliable, proven solution that is currently being used in these scenarios. First and foremost, Berkeley DB (or BDB for short) is very very fast. It can process tens or hundreds of thousands of transactions per second. It can be used as a pure in-memory database, or as a disk-persistent database. BDB provides high availability – if one board in the router fails, the system can automatically failover to another board – no manual intervention required. BDB is self-administering – there’s no need for manual intervention in order to maintain a BDB application. No need to send a technician to a remote site in the middle of nowhere on a freezing winter day to perform maintenance operations. BDB is used in over 200 million deployments worldwide for the past two decades for mission-critical applications such as the one described here. You have a choice of spending valuable resources to implement similar functionality, or, you could simply embed BDB in your application and off you go! I know what I’d do – choose BDB, so I can focus on my business problem. What will you do? /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin-top:0in; mso-para-margin-right:0in; mso-para-margin-bottom:10.0pt; mso-para-margin-left:0in; line-height:115%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin;}

Read the article

A Basic Thread

- by Joe Mayo

Most of the programs written are single-threaded, meaning that they run on the main execution thread. For various reasons such as performance, scalability, and/or responsiveness additional threads can be useful. .NET has extensive threading support, from the basic threads introduced in v1.0 to the Task Parallel Library (TPL) introduced in v4.0. To get started with threads, it's helpful to begin with the basics; starting a Thread. Why Do I Care? The scenario I'll use for needing to use a thread is writing to a file. Sometimes, writing to a file takes a while and you don't want your user interface to lock up until the file write is done. In other words, you want the application to be responsive to the user. How Would I Go About It? The solution is to launch a new thread that performs the file write, allowing the main thread to return to the user right away. Whenever the file writing thread completes, it will let the user know. In the meantime, the user is free to interact with the program for other tasks. The following examples demonstrate how to do this. Show Me the Code? The code we'll use to work with threads is in the System.Threading namespace, so you'll need the following using directive at the top of the file: using System.Threading; When you run code on a thread, the code is specified via a method. Here's the code that will execute on the thread: private static void WriteFile() { Thread.Sleep(1000); Console.WriteLine("File Written."); } The call to Thread.Sleep(1000) delays thread execution. The parameter is specified in milliseconds, and 1000 means that this will cause the program to sleep for approximately 1 second. This method happens to be static, but that's just part of this example, which you'll see is launched from the static Main method. A thread could be instance or static. Notice that the method does not have parameters and does not have a return type. As you know, the way to refer to a method is via a delegate. There is a delegate named ThreadStart in System.Threading that refers to a method without parameters or return type, shown below: ThreadStart fileWriterHandlerDelegate = new ThreadStart(WriteFile); I'll show you the whole program below, but the ThreadStart instance above goes in the Main method. The thread uses the ThreadStart instance, fileWriterHandlerDelegate, to specify the method to execute on the thread: Thread fileWriter = new Thread(fileWriterHandlerDelegate); As shown above, the argument type for the Thread constructor is the ThreadStart delegate type. The fileWriterHandlerDelegate argument is an instance of the ThreadStart delegate type. This creates an instance of a thread and what code will execute, but the new thread instance, fileWriter, isn't running yet. You have to explicitly start it, like this: fileWriter.Start(); Now, the code in the WriteFile method is executing on a separate thread. Meanwhile, the main thread that started the fileWriter thread continues on it's own. You have two threads running at the same time. Okay, I'm Starting to Get Glassy Eyed. How Does it All Fit Together? The example below is the whole program, pulling all the previous bits together. It's followed by its output and an explanation. using System; using System.Threading; namespace BasicThread { class Program { static void Main() { ThreadStart fileWriterHandlerDelegate = new ThreadStart(WriteFile); Thread fileWriter = new Thread(fileWriterHandlerDelegate); Console.WriteLine("Starting FileWriter"); fileWriter.Start(); Console.WriteLine("Called FileWriter"); Console.ReadKey(); } private static void WriteFile() { Thread.Sleep(1000); Console.WriteLine("File Written"); } } } And here's the output: Starting FileWriter Called FileWriter File Written So, Why are the Printouts Backwards? The output above corresponds to Console.Writeline statements in the program, with the second and third seemingly reversed. In a single-threaded program, "File Written" would print before "Called FileWriter". However, this is a multi-threaded (2 or more threads) program. In multi-threading, you can't make any assumptions about when a given thread will run. In this case, I added the Sleep statement to the WriteFile method to greatly increase the chances that the message from the main thread will print first. Without the Thread.Sleep, you could run this on a system with multiple cores and/or multiple processors and potentially get different results each time. Interesting Tangent but What Should I Get Out of All This? Going back to the main point, launching the WriteFile method on a separate thread made the program more responsive. The file writing logic ran for a while, but the main thread returned to the user, as demonstrated by the print out of "Called FileWriter". When the file write finished, it let the user know via another print statement. This was a very efficient use of CPU resources that made for a more pleasant user experience. Joe

Read the article

7 Steps To Cut Recruiting Costs & Drive Exceptional Business Results

- by Oracle Accelerate for Midsize Companies

By Steve Viarengo, Vice President Product Management, Oracle Taleo Cloud Services Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 In good times, trimming operational costs is an ongoing goal. In tough times, it’s a necessity. In both good times and bad, however, recruiting occurs. Growth increases headcount in good times, and opportunistic or replacement hiring occurs in slow business cycles. By employing creative recruiting strategies in tandem with the latest technology developments, you can reduce recruiting costs while driving exceptional business results. Here are some critical areas to focus on. 1. Target Direct Cost Savings Total recruiting process expenses are the sum of external costs plus internal labor costs. Most organizations can reduce recruiting expenses with direct cost savings. While additional savings on indirect costs can be realized from process improvement and efficiency gains, there are direct cost savings and benefits readily available in three broad areas: sourcing, assessments, and green recruiting. 2. Sourcing: Reduce Agency Costs Agency search firm fees can amount to 35 percent of a new employee’s annual base salary. Typically taken from the hiring department budget, these fees may not be visible to HR. By relying on internal mobility programs, referrals, candidate pipelines, and corporate career Websites, organizations can reduce or eliminate this agency spend. And when you do have to pay third-party agency fees, you can optimize the value you receive by collaborating with agencies to identify referred candidates, ensure access to candidate data and history, and receive automatic notifications and correspondence. 3. Sourcing: Reduce Advertising Costs You can realize significant cost reductions by placing all job positions on your corporate career Website. This will allow you to reap a substantial number of candidates at minimal cost compared to job boards and other sourcing options. 4. Sourcing: Internal Talent Pool Internal talent pools provide a way to reduce sourcing and advertising costs while delivering improved productivity and retention. Internal redeployment reduces costs and ramp-up time while increasing retention and employee satisfaction. 5. Sourcing: External Talent Pool Strategic recruiting requires identifying and matching people with a given set of skills to a particular job while efficiently allocating sourcing expenditures. By using an e-recruiting system (which drives external talent pool management) with a candidate relationship database, you can automate prescreening and candidate matching while communicating with targeted candidates. Candidate relationship management can lower sourcing costs by marketing new job opportunities to candidates sourced in the past. By mining the talent pool in this fashion, you eliminate the need to source a new pool of candidates for each new requisition. Managing and mining the corporate candidate database can reduce the sourcing cost per candidate by as much as 50 percent. 6. Assessments: Reduce Turnover Costs By taking advantage of assessments during the recruitment process, you can achieve a range of benefits, including better productivity, superior candidate performance, and lower turnover (providing considerable savings). Assessments also save recruiter and hiring manager time by focusing on a short list of qualified candidates. Hired for fit, such candidates tend to stay with the organization and produce quality work—ultimately driving revenue. 7. Green Recruiting: Reduce Paper and Processing Costs You can reduce recruiting costs by automating the process—and making it green. A paperless process informs candidates that you’re dedicated to green recruiting. It also leads to direct cost savings. E-recruiting reduces energy use and pollution associated with manufacturing, transporting, and recycling paper products. And process automation saves energy in mailing, storage, handling, filing, and reporting tasks. Direct cost savings come from reduced paperwork related to résumés, advertising, and onboarding. Improving the recruiting process through sourcing, assessments, and green recruiting not only saves costs. It also positions the company to improve the talent base during the recession while retaining the ability to grow appropriately in recovery. /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Calibri","sans-serif"; mso-bidi-font-family:"Times New Roman";} Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Calibri","sans-serif"; mso-bidi-font-family:"Times New Roman";}

Read the article

EM CLI, diving in and beyond!

- by Maureen Byrne

v\:* {behavior:url(#default#VML);} o\:* {behavior:url(#default#VML);} w\:* {behavior:url(#default#VML);} .shape {behavior:url(#default#VML);} Doing more in less time… Isn’t that what we all strive to do? With this in mind, I put together two screen watches on Oracle Enterprise Manager 12c command line interface, or EM CLI as it is also known. There is a wealth of information on any topic that you choose to read about, from manual pages to coding documents…might I even say blog posts? In our busy lives it is so nice to just sit back with a short video, watch and learn enough to dive in. Doing more in less time, is the essence of EM CLI. It enables you to script fundamental and complex administrative tasks in an elegant way, thanks to the Jython scripting language. Repetitive tasks can be scripted and reused again and again. Sure, a Graphical User Interface provides a more intuitive step by step approach to tasks, and it provides a way of quickly becoming familiar with a product and its many features, and it is definitely the way to go when viewing performance data and historical trending…but for repetitive and complex tasks, scripting is the way to go! Lets us take the everyday task of creating an administrator. Using EM CLI in interactive mode the command could look like this.. emcli>create_user(name='jan.doe', type='EXTERNAL_USER') This command creates an administrator called jan.doe which is an externally authenticated user, possibly LDAP or SSO, defined by the EXTERNAL_USER tag. The create_user procedure takes many arguments; see the documentation for more information. Now, where EM CLI really shines and shows power is in creating multiple users. Regardless of the number, tens or thousands, the effort is the same. With the use of a standard programming construct, a loop, you can place your create_user() procedure within it. Using a loop allows you to iterate through a previously created list, creating new users until the list is complete. Using EM CLI in Script mode, your Jython loop would look something like this… for user in list_of_users: create_user(name=user, expire=’true’, password=’welcome123’) This Jython code snippet iterates through a previously defined list of names, list_of_users, and iterates through the list, taking each name, user in this case, and creates an administrator sets the password to welcome123, but forces the user to reset it when they first login. This is only one of over four hundred procedures created to expose Oracle Enterprise Manager 12c functionality in a powerful and programmatic way. It is a few months since we released EM CLI with scripting option. We are seeing many users adapt to this fun and powerful way of using Oracle Enterprise Manager 12c. What are the first steps? Watch these screen watches, and dive in. The first screen watch steps you through where and how to download and install and how to run your first few commands. The Second screen watch steps you through a few scripts. Next time, I am going to show you the basic building blocks to writing a Jython script to perform Oracle Enterprise Manager 12c administrative tasks. Join this growing group of EM CLI users…. Dive in! Normal 0 false false false false EN-US X-NONE X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin-top:0in; mso-para-margin-right:0in; mso-para-margin-bottom:10.0pt; mso-para-margin-left:0in; line-height:115%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi;}

Read the article

Most Innovative IDM Projects: Awards at OpenWorld

- by Tanu Sood

On Tuesday at Oracle OpenWorld 2012, Oracle recognized the winners of Innovation Awards 2012 at a ceremony presided over by Hasan Rizvi, Executive Vice President at Oracle. Oracle Fusion Middleware Innovation Awards recognize customers for achieving significant business value through innovative uses of Oracle Fusion Middleware offerings. Winners are selected based on the uniqueness of their business case, business benefits, level of impact relative to the size of the organization, complexity and magnitude of implementation, and the originality of architecture. This year’s Award honors customers for their cutting-edge solutions driving business innovation and IT modernization using Oracle Fusion Middleware. The program has grown over the past 6 years, receiving a record number of nominations from customers around the globe. The winners were selected by a panel of judges that ranked each nomination across multiple different scoring categories. Congratulations to both Avea and ETS for winning this year’s Innovation Award for Identity Management. Identity Management Innovation Award 2012 Winner – Avea Company: Founded in 2004, AveA is the sole GSM 1800 mobile operator of Turkey and has reached a nationwide customer base of 12.8 million as of the end of 2011 Region: Turkey (EMEA) Products: Oracle Identity Manager, Oracle Identity Analytics, Oracle Access Management Suite Business Drivers: · To manage the agility and scale required for GSM Operations and enable call center efficiency by enabling agents to change their identity profiles (accounts and entitlements) rapidly based on call load. · Enhance user productivity and call center efficiency with self service password resets · Enforce compliance and audit reporting · Seamless identity management between AveA and parent company Turk Telecom Innovation and Results: · One of the first Sun2Oracle identity management migrations designed for high performance provisioning and trusted reconciliation built with connectors developed on the ICF architecture that provides custom user interfaces for dynamic and rapid management of roles and entitlements along with entitlement level attestation using closed loop remediation between Oracle Identity Manager and Oracle Identity Analytics. · Dramatic reduction in identity administration and call center password reset tasks leading to 20% reduction in administration costs and 95% reduction in password related calls. · Enhanced user productivity by up to 25% to date · Enforced enterprise security and reduced risk · Cost-effective compliance management · Looking to seamlessly integrate with parent and sister companies’ infrastructure securely. Identity Management Innovation Award 2012 Winner – Education Testing Service (ETS) See last year's winners here --Company: ETS is a private nonprofit organization devoted to educational measurement and research, primarily through testing. Region: U.S.A (North America) Products: Oracle Access Manager, Oracle Identity Federation, Oracle Identity Manager Business Drivers: ETS develops and administers more than 50 million achievement and admissions tests each year in more than 180 countries, at more than 9,000 locations worldwide. As the business becomes more globally based, having a robust solution to security and user management issues becomes paramount. The organizations was looking for: · Simplified user experience for over 3000 company users and more than 6 million dynamic student and staff population · Infrastructure and administration cost reduction · Managing security risk by controlling 3rd party access to ETS systems · Enforce compliance and manage audit reporting · Automate on-boarding and decommissioning of user account to improve security, reduce administration costs and enhance user productivity · Improve user experience with simplified sign-on and user self service Innovation and Results: 1. Manage Risk · Centralized system to control user access · Provided secure way of accessing service providers' application using federated SSO. · Provides reporting capability for auditing, governance and compliance. 2. Improve efficiency · Real-Time provisioning to target systems · Centralized provisioning system for user management and access controls. · Enabling user self services. 3. Reduce cost · Re-using common shared services for provisioning, SSO, Access by application reducing development cost and time. · Reducing infrastructure and maintenance cost by decommissioning legacy/redundant IDM services. · Reducing time and effort to implement security functionality in business applications (“onboard” instead of new development). ETS was able to fold in new and evolving requirement in addition to the initial stated goals realizing quick ROI and successfully meeting business objectives. Congratulations to the winners once again. We will be sure to bring you more from these Innovation Award winners over the next few months.

Read the article

Oracle NoSQL Database Exceeds 1 Million Mixed YCSB Ops/Sec

- by Charles Lamb

We ran a set of YCSB performance tests on Oracle NoSQL Database using SSD cards and Intel Xeon E5-2690 CPUs with the goal of achieving 1M mixed ops/sec on a 95% read / 5% update workload. We used the standard YCSB parameters: 13 byte keys and 1KB data size (1,102 bytes after serialization). The maximum database size was 2 billion records, or approximately 2 TB of data. We sized the shards to ensure that this was not an "in-memory" test (i.e. the data portion of the B-Trees did not fit into memory). All updates were durable and used the "simple majority" replica ack policy, effectively 'committing to the network'. All read operations used the Consistency.NONE_REQUIRED parameter allowing reads to be performed on any replica. In the past we have achieved 100K ops/sec using SSD cards on a single shard cluster (replication factor 3) so for this test we used 10 shards on 15 Storage Nodes with each SN carrying 2 Rep Nodes and each RN assigned to its own SSD card. After correcting a scaling problem in YCSB, we blew past the 1M ops/sec mark with 8 shards and proceeded to hit 1.2M ops/sec with 10 shards. Hardware Configuration We used 15 servers, each configured with two 335 GB SSD cards. We did not have homogeneous CPUs across all 15 servers available to us so 12 of the 15 were Xeon E5-2690, 2.9 GHz, 2 sockets, 32 threads, 193 GB RAM, and the other 3 were Xeon E5-2680, 2.7 GHz, 2 sockets, 32 threads, 193 GB RAM. There might have been some upside in having all 15 machines configured with the faster CPU, but since CPU was not the limiting factor we don't believe the improvement would be significant. The client machines were Xeon X5670, 2.93 GHz, 2 sockets, 24 threads, 96 GB RAM. Although the clients had 96 GB of RAM, neither the NoSQL Database or YCSB clients require anywhere near that amount of memory and the test could have just easily been run with much less. Networking was all 10GigE. YCSB Scaling Problem We made three modifications to the YCSB benchmark. The first was to allow the test to accommodate more than 2 billion records (effectively int's vs long's). To keep the key size constant, we changed the code to use base 32 for the user ids. The second change involved to the way we run the YCSB client in order to make the test itself horizontally scalable.The basic problem has to do with the way the YCSB test creates its Zipfian distribution of keys which is intended to model "real" loads by generating clusters of key collisions. Unfortunately, the percentage of collisions on the most contentious keys remains the same even as the number of keys in the database increases. As we scale up the load, the number of collisions on those keys increases as well, eventually exceeding the capacity of the single server used for a given key.This is not a workload that is realistic or amenable to horizontal scaling. YCSB does provide alternate key distribution algorithms so this is not a shortcoming of YCSB in general. We decided that a better model would be for the key collisions to be limited to a given YCSB client process. That way, as additional YCSB client processes (i.e. additional load) are added, they each maintain the same number of collisions they encounter themselves, but do not increase the number of collisions on a single key in the entire store. We added client processes proportionally to the number of records in the database (and therefore the number of shards). This change to the use of YCSB better models a use case where new groups of users are likely to access either just their own entries, or entries within their own subgroups, rather than all users showing the same interest in a single global collection of keys. If an application finds every user having the same likelihood of wanting to modify a single global key, that application has no real hope of getting horizontal scaling. Finally, we used read/modify/write (also known as "Compare And Set") style updates during the mixed phase. This uses versioned operations to make sure that no updates are lost. This mode of operation provides better application behavior than the way we have typically run YCSB in the past, and is only practical at scale because we eliminated the shared key collision hotspots.It is also a more realistic testing scenario. To reiterate, all updates used a simple majority replica ack policy making them durable. Scalability Results In the table below, the "KVS Size" column is the number of records with the number of shards and the replication factor. Hence, the first row indicates 400m total records in the NoSQL Database (KV Store), 2 shards, and a replication factor of 3. The "Clients" column indicates the number of YCSB client processes. "Threads" is the number of threads per process with the total number of threads. Hence, 90 threads per YCSB process for a total of 360 threads. The client processes were distributed across 10 client machines. Shards KVS Size Clients Mixed (records) Threads OverallThroughput(ops/sec) Read Latencyav/95%/99%(ms) Write Latencyav/95%/99%(ms) 2 400m(2x3) 4 90(360) 302,152 0.76/1/3 3.08/8/35 4 800m(4x3) 8 90(720) 558,569 0.79/1/4 3.82/16/45 8 1600m(8x3) 16 90(1440) 1,028,868 0.85/2/5 4.29/21/51 10 2000m(10x3) 20 90(1800) 1,244,550 0.88/2/6 4.47/23/53

Read the article

Webcor Builders Coordinates Construction Schedules and Mitigates Potential Delays More Efficiently with Integrated Project Management

- by Sylvie MacKenzie, PMP

Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman","serif";} With more than 40 years of commercial construction experience, Webcor Builders is a leading builder of distinguished, high-profile projects, including high-rise condominiums and hotels, laboratories, healthcare centers, and public works projects. Webcor is also known for its award-winning concrete, interior construction, historic restoration, and seismic renovation work. The company has completed more than 50 million square feet of projects to date. Considering the variety and complexity of the construction projects Webcor undertakes, an integrated project management solution is critical to ensuring optimal efficiency and completing client projects on time and on budget. The company previously used a number of scheduling systems for its various building projects. These packages provided different levels of schedule detail and required schedulers, engineers, and other employees to learn multiple systems. From an IT cost and complexity perspective, the company had to manage multiple scheduling systems and pay for multiple sets of licenses. The company looked to standardize on an enterprise project management system, and selected Oracle’s Primavera P6 Enterprise Project Portfolio Management. Webcor uses the solution’s advanced capabilities to schedule complex projects, analyze delays, model and propose multiple scenarios to demonstrate and mitigate delays and cost overruns, and process that information efficiently to deliver the scheduling precision that public and private projects require. In fact, the solution was instrumental in helping the company’s expansion into public sector projects during the recent economic downturn, and with Primavera P6 in place, it can deliver the precise schedule reporting required for large public projects. With Primavera P6 in place, the company could deliver the precise scheduling and milestone reporting capabilities required for large public projects. The solution is in managing the high-profile University of California – Berkeley Memorial Stadium project. Webcor was hired as construction manager and general contractor for the stadium renovation project, which is a fast-paced project located near the seismically active Hayward Fault Zone. Due to the University of California’s football schedule, meeting the Universities deadline for the coming season placed Webcor in a situation where risk awareness and early warnings of issues would be paramount. Webcor and the extended project team needed a solution that could instantly analyze alternate scenarios to mitigate potential delays; Primavera would deliver those answers.The team would also need to enable multiple stakeholders to use an internet-based platform to access the schedule from various locations, and model complicated sequencing requirements where swift decisions would be made to keep the project on track. The schedule is an integral part of Webcor’s construction management process for the stadium project. Rather than providing the client with the industry-standard monthly update, Webcor updates the critical path method (CPM) schedule on a weekly basis. The project team also reviews the schedule and updates weekly to confirm that progress and forecasted performance are accurate. Hired by the University for their ability to deliver in high risk environments The Webcor team was hit recently with a design supplement that could have added up to 70 days to the project. Using Oracle Primavera P6 the team sprung into action analyzing multiple “what if” scenarios to review mitigation means and methods. Determined to make sure the Bears could take the field in the coming season the project team nearly eliminated the impact with their creative analysis in working the schedule. The total time from the issuance of the final design supplement to an agreed mitigation response was less than one week; leveraging the Oracle Primavera solution Webcor was able to deliver superior customer value With the ability to efficiently manage projects and schedules, Webcor can ensure it completes its projects on time and on budget, as well as inform clients about what changes to plans will mean in terms of delays and additional costs. Read the complete customer case study at : http://www.oracle.com/us/corporate/customers/customersearch/webcor-builders-1-primavera-ss-1639886.html

Read the article

Why everybody should do Sales!

- by FelixWehmeyer

I speak with many business students and ask them what job they want to get into. Most of them tell me they want a job in Marketing, Management Consulting or Finance. I hardly ever hear “Sales, that is what I want to do”, and I often wonder why. I would like to start with a quote from Zig Ziglar, a successful salesman: "Nothing happens until someone sells something." But to get back to the main point, why wouldn’t you want to get in sales? When people think of sales, they picture a typical salesman in their head and think that selling is scary and all about manipulating, pressuring and pushing someone into buying something they don’t need. Are these stereotypes accurate? I don’t believe so: So why should you want to be in sales? If you think about selling as providing the solution for the problem and talking about the benefits of making a decision, then every job in this world comes out of selling. In every job you deal with coworkers that you want to convince of your ideas or convincing your boss that the project you want to work on is good for the company. These days, consumers and businesses are very well informed about services and products. When we are talking about highly complex products, such as IT solutions, businesses don’t accept your run-of-the-mill salesman who is pushing a sale. These are often long projects where salespeople have a consulting and leading role. Salespeople need to be able to consult companies and customers with their problem and convince a client that their solution is the best fit. Next to the fact that sales, is by far, not as scary and shady as you thought, there are a few points that will make you want to consider a sales career: Negotiating skills – When you are in sales you will learn how to negotiate. Salespeople learn to listen to their customers and try to make them happy, overcoming objections and come to a final agreement that both parties are happy with. Persistence/Challenge – As a salesperson you will often hear a negative answer, in a sales role you will start to embrace this and see a ‘no’ as a challenge not as a rejection. This attitude change can help you a lot in your career, but also in your personal life. You will become more optimistic and gain a go-getter attitude. Salary – As salespeople are seen as the moneymakers for the company, companies often reward their sales teams generously. Most likely in a sales role, you will receive a good basic salary and often you get nice bonuses on top of that based on your performance. Oracle is, for instance, the company that offers the highest average commission in the world. Further you can expect many other benefits as companies know that there is a high demand for good salespeople. Teamwork – Sales is a lot like having your own business, you are responsible for your own territory or set of clients. You are the one who is responsible for the revenue coming from that territory. So in order to gain revenue you will have to work together with many departments and people to make that happen. Every (potential) client could be seen as a different project, and you are the project leader. Understanding customers and the business – From any job that you choose sales will get you the most insight in the market. Salespeople are usually well-connected, talk with different customers and learn about the market and are up-to-date about all latest changes. Even if you want to change to a different role in the long run, you have a great head start as you understand the market and customers like no one else. Job security – Look at all the job postings out there. Many of them are sales-related. So if you want to have a steady job, plenty of choice and companies willing to invest in you, sales could be something for you. Are you interested in exploring a sales career? At Oracle we are always looking for good sales professionals and fresh graduates who want to get into sales! For many languages such as Flemish, Dutch, German, French, Swedish and Norwegian (and more) we are currently looking for graduates who want to develop their career in Oracle. Please have a look at this article for the experience of a Business Development Consultant at Oracle in Dublin. Want to learn more about this job check out this link or send an email to jessica.ebbelaar-at-oracle.com! Have a look at our website http://campus.oracle.com for all of our other latest sales and non-sales vacancies!

Read the article

What Counts for a DBA: Humility

- by drsql

In football (the American sort, naturally,) there are a select group of players who really hope to never have their names called during the game. They are members of the offensive line, and their job is to protect other players so they can deliver the ball to the goal to score points. When you do hear their name called, it is usually because they made a mistake and the player that they were supposed to protect ended up flat on his back admiring the clouds in the sky instead of advancing towards the goal to scoring point. Even on the rare occasion their name is called for a good reason, it is usually because they were making up for a teammate who had made a mistake and they covered up for them. The role of offensive lineman is a very good analogy for the role of the admin DBA. As a DBA, you are called on to be barely visible and rarely heard, protecting the company data assets tenaciously, even though the enemies to our craft surround us on all sides:. Developers: Cries of ‘foul!’ often ensue when the DBA says that they want data integrity to be stringently enforced and that documentation is needed so they can support systems, mostly because every error occurrence in the enterprise will be initially blamed on the database and fall to the DBA to troubleshoot. Insisting too loudly may bring those cries of ‘foul’ that somewhat remind you of when your 2 year old daughter didn't want to go to bed. The result of this petulance is that the next "enemy" gets involved. Managers: The concerns that motivate DBAs to argue will not excite the kind of manager who gets his technical knowledge from a glossy magazine filled with buzzwords, charts, and pretty pictures. However, the other programmers in the organization will tickle the buzzword void with a stream of new-sounding ideas and technologies constantly, along with warnings that if we did care about data integrity and document things, the budget would explode! In contrast, the arguments for integrity of data and supportability tend to be about as exciting as watching grass grow, and far too many manager types seem to prefer to smoke it than watch it. Packaged Applications: The DBA is rarely given a chance to review a new application that is being demonstrated for the enterprise, and rarer still is the DBA that gets a veto of an application because the database it uses has clearly been created by an architect that won't read a data modeling book because he is already married. More often than not this leads to hours of work for the DBA trying to performance-tune a database with a menagerie of rules that must be followed to stay within the application support agreement, such as no changing indexes on a third party schema even though there are 10 billion rows instead of the 10 thousand when the system was last optimized. Hardware Failures: Physical disks, networking devices, memory, and backup devices all come with a measure known as ‘mean time before failure’ and it is never listed in centuries or eons. More like years, and the term ‘mean’ indicates that half of the devices are expected to fail before that, which by my calendar means any hour of any day that it wants to fail it will. But the DBA sucks it up and does the task at hand with a humility that makes them nearly invisible to all but the most observant person in the organization. The best DBAs I know are so proactive in their relentless pursuit of perfection that they detect many of the bugs (which they seldom caused) in the system well before they become a problem. In the end the DBA gets noticed for one of same two reasons as the offensive lineman. You make a mistake, like dropping a critical production database that had never been backed up; or when a system crashes for any reason whatsoever and they are on the spot with troubleshooting and system restoration plans that have been well thought out, tested, and tested again. Not because there is any glory in it, but because it is what they do. Note: The characteristics of the professions referred to in this blog are meant to be overstated stereotypes for humorous effect, and even some DBAs aren't quite this perfect. If you are reading this far and haven’t hand written a 10 page flaming comment about how you are a _______ and you aren’t like this, that is awesome. Not every situation applies to everyone, but if you have never worked with a bad packaged app, a magazine trained manager, programmers that aren’t team players, or hardware that occasionally failed, relax and go have a unicorn sandwich before you wake up.

Read the article

how to label a cuboid using open gl?

- by usha

hi this is how my 3dcuboid looks ,i have attached complete code , i want to label this cuboid using different name across sides how is it possible using opengl in android...plz help me out public class MyGLRenderer implements Renderer { Context context; Cuboid rect; private float mCubeRotation; // private static float angleCube = 0; // Rotational angle in degree for cube (NEW) // private static float speedCube = -1.5f; // Rotational speed for cube (NEW) public MyGLRenderer(Context context) { rect = new Cuboid(); this.context = context; } public void onDrawFrame(GL10 gl) { // TODO Auto-generated method stub gl.glClear(GL10.GL_COLOR_BUFFER_BIT | GL10.GL_DEPTH_BUFFER_BIT); gl.glLoadIdentity(); // Reset the model-view matrix gl.glTranslatef(0.2f, 0.0f, -8.0f); // Translate right and into the screen gl.glScalef(0.8f, 0.8f, 0.8f); // Scale down (NEW) gl.glRotatef(mCubeRotation, 1.0f, 1.0f, 1.0f); // gl.glRotatef(angleCube, 1.0f, 1.0f, 1.0f); // rotate about the axis (1,1,1) (NEW) rect.draw(gl); mCubeRotation -= 0.15f; //angleCube += speedCube; } public void onSurfaceChanged(GL10 gl, int width, int height) { // TODO Auto-generated method stub if (height == 0) height = 1; // To prevent divide by zero float aspect = (float)width / height; // Set the viewport (display area) to cover the entire window gl.glViewport(0, 0, width, height); // Setup perspective projection, with aspect ratio matches viewport gl.glMatrixMode(GL10.GL_PROJECTION); // Select projection matrix gl.glLoadIdentity(); // Reset projection matrix // Use perspective projection GLU.gluPerspective(gl, 45, aspect, 0.1f, 100.f); gl.glMatrixMode(GL10.GL_MODELVIEW); // Select model-view matrix gl.glLoadIdentity(); // Reset } public void onSurfaceCreated(GL10 gl, EGLConfig config) { // TODO Auto-generated method stub gl.glClearColor(0.0f, 0.0f, 0.0f, 1.0f); // Set color's clear-value to black gl.glClearDepthf(1.0f); // Set depth's clear-value to farthest gl.glEnable(GL10.GL_DEPTH_TEST); // Enables depth-buffer for hidden surface removal gl.glDepthFunc(GL10.GL_LEQUAL); // The type of depth testing to do gl.glHint(GL10.GL_PERSPECTIVE_CORRECTION_HINT, GL10.GL_NICEST); // nice perspective view gl.glShadeModel(GL10.GL_SMOOTH); // Enable smooth shading of color gl.glDisable(GL10.GL_DITHER); // Disable dithering for better performance }} public class Cuboid{ private FloatBuffer mVertexBuffer; private FloatBuffer mColorBuffer; private ByteBuffer mIndexBuffer; private float vertices[] = { //width,height,depth -2.5f, -1.0f, -1.0f, 1.0f, -1.0f, -1.0f, 1.0f, 1.0f, -1.0f, -2.5f, 1.0f, -1.0f, -2.5f, -1.0f, 1.0f, 1.0f, -1.0f, 1.0f, 1.0f, 1.0f, 1.0f, -2.5f, 1.0f, 1.0f }; private float colors[] = { // R,G,B,A COLOR 0.0f, 1.0f, 0.0f, 1.0f, 0.0f, 1.0f, 0.0f, 1.0f, 1.0f, 0.5f, 0.0f, 1.0f, 1.0f, 0.5f, 0.0f, 1.0f, 1.0f, 0.0f, 0.0f, 1.0f, 1.0f, 0.0f, 0.0f, 1.0f, 0.0f, 0.0f, 1.0f, 1.0f, 1.0f, 0.0f, 1.0f, 1.0f }; private byte indices[] = { // VERTEX 0,1,2,3,4,5,6,7 REPRESENTATION FOR FACES 0, 4, 5, 0, 5, 1, 1, 5, 6, 1, 6, 2, 2, 6, 7, 2, 7, 3, 3, 7, 4, 3, 4, 0, 4, 7, 6, 4, 6, 5, 3, 0, 1, 3, 1, 2 }; public Cuboid() { ByteBuffer byteBuf = ByteBuffer.allocateDirect(vertices.length * 4); byteBuf.order(ByteOrder.nativeOrder()); mVertexBuffer = byteBuf.asFloatBuffer(); mVertexBuffer.put(vertices); mVertexBuffer.position(0); byteBuf = ByteBuffer.allocateDirect(colors.length * 4); byteBuf.order(ByteOrder.nativeOrder()); mColorBuffer = byteBuf.asFloatBuffer(); mColorBuffer.put(colors); mColorBuffer.position(0); mIndexBuffer = ByteBuffer.allocateDirect(indices.length); mIndexBuffer.put(indices); mIndexBuffer.position(0); } public void draw(GL10 gl) { gl.glFrontFace(GL10.GL_CW); gl.glVertexPointer(3, GL10.GL_FLOAT, 0, mVertexBuffer); gl.glColorPointer(4, GL10.GL_FLOAT, 0, mColorBuffer); gl.glEnableClientState(GL10.GL_VERTEX_ARRAY); gl.glEnableClientState(GL10.GL_COLOR_ARRAY); gl.glDrawElements(GL10.GL_TRIANGLES, 36, GL10.GL_UNSIGNED_BYTE, mIndexBuffer); gl.glDisableClientState(GL10.GL_VERTEX_ARRAY); gl.glDisableClientState(GL10.GL_COLOR_ARRAY); } } public class Draw3drect extends Activity { private GLSurfaceView glView; // Use GLSurfaceView // Call back when the activity is started, to initialize the view @Override protected void onCreate(Bundle savedInstanceState) { super.onCreate(savedInstanceState); glView = new GLSurfaceView(this); // Allocate a GLSurfaceView glView.setRenderer(new MyGLRenderer(this)); // Use a custom renderer this.setContentView(glView); // This activity sets to GLSurfaceView } // Call back when the activity is going into the background @Override protected void onPause() { super.onPause(); glView.onPause(); } // Call back after onPause() @Override protected void onResume() { super.onResume(); glView.onResume(); } }

Read the article

WEBLOGIC 12C HANDS-ON BOOTCAMP

- by agallego

Oracle PartnerNetwork | Account | Feedback JOIN THE ORACLE WEBLOGIC PARTNER COMMUNITY AND ATTEND A WEBLOGIC 12C HANDS-ON BOOTCAMP Dear partner As a valued partner we would like to invite you for the WebLogic Partner Community and our WebLogic 12c hands-on Bootcamps – free of charge! Please first login at http://partner.oracle.com and then visit: WebLogic Partner Community. (If you need support with your account please contact the Oracle Partner Business Center). The goal of the WebLogic Partner Community is to provide you with the latest information on Oracle's offerings and to facilitate the exchange of experience within community members. Register Now FREE Assessment vouchers to become certified and WebLogic Server 12c 200 new Features and Training Connect and Network WebLogic Blogs WebLogic on Facebook WebLogic on LinkedIn WebLogic on Twitter WebLogic on Oracle Mix WebLogic 12c hands-on Workshops We offer free3 days hands-on WebLogic 12c workshops for Oracle partners who want to become Application Grid Specialized: Register Here! Country Date Location Registration Germany 3-5 April 2012 Oracle Düsseldorf Click here France 24-26 April 2012 Oracle Colombes Click here Spain 08-10 May 2012 Oracle Madrid Click here Netherlands 22-24 May 2012 Oracle Amsterdam Click here United Kingdom 06-08 June 2012 Oracle Reading Click here Italy 19-21 June 2012 Oracle Cinisello Balsamo Click here Portugal 10-12 July 2012 Oracle Lisbon Click here Skill requirements Attendees need to have the following skills as this is required by the product-set and to make sure they get the most out of the training: Basic knowledge in Java and JavaEE Understanding the Application Server concept Basic knowledge in older releases of WebLogic Server would be beneficial Member of WebLogic Partner Community for registration please vist http://www.oracle.com/partners/goto/wls-emea Hardware requirements Every participant works on his own notebook. The minimal hardware requirements are: 4Gb physical RAM (we will boot the image with 2Gb RAM) dual core CPU 15 GB HD Software requirements Please install Oracle VM VirtualBox 4.1.8 Follow-up and certification With the workshop registration you agree to the following next steps Follow-up training attend and pass the Oracle Application Grid Certified Implementation Specialist Registration For details and registration please visit Register Here Free WebLogic Certification (Free assessment voucher to become certified) For all WebLogic experts, we offer free vouchers worth $195 for the Oracle Application Grid Certified Implementation Specialist assessment. To demonstrate your WebLogic knowledge you first have to pass the free online assessment Oracle Application Grid PreSales Specialist. For free vouchers, please send an e-mail with the screenshot of your Oracle Application Grid PreSales certirficate to [email protected] including your Name, Company, E-mail and Country. Note: This offer is limited to partners from Europe Middle East and Africa. Partners from other countries please contact your Oracle partner manager. WebLogic Specialization To become specialized in Application Grid, please make sure that you access the: Application Grid Specialization Guide Application Grid Specialization Checklist If you have any questions please contact the Oracle Partner Business Center. Oracle WebLogic Server 12c Key New Capabilities Java EE 6 and Developer Productivity Simplified Deployment and Management with Virtualization Integrated Traffic Management Enhanced High Availability and Disaster Recovery Much Higher Performance For more information please visit: Presentation from the WebLogic 12c launch Technical Presentation from the WebLogic 12c launch WebLogic OTN Website WebLogic 12c Virtual Conference Environment WebLogic Partner Community For regular information become a member in the WebLogic Partner Community please visit: http://www.oracle.com/partners/goto/wls-emea (OPN account required). If you need support with your account please contact the Oracle Partner Business Center. Best regards, Jürgen Kress WebLogic Partner Adoption EMEA Tel. +49 89 1430 1479 E-Mail: [email protected] Copyright © 2012, Oracle and/or its affiliates. All rights reserved. Contact PBC | Legal Notices and Terms of Use | Privacy Statement Oracle Corporation - Worldwide Headquarters, 500 Oracle Parkway, OPL - E-mail Services, Redwood Shores, CA 94065, United States Your privacy is important to us. You can login to your account to update your e-mail subscriptions or you can opt-out of all Oracle Marketing e-mails at any time. Please note that opting-out of Marketing communications does not affect your receipt of important business communications related to your current relationship with Oracle such as Security Updates, Event Registration notices, Account Management and Support/Service communications.

Read the article

Using R to Analyze G1GC Log Files

- by user12620111

Using R to Analyze G1GC Log Files body, td { font-family: sans-serif; background-color: white; font-size: 12px; margin: 8px; } tt, code, pre { font-family: 'DejaVu Sans Mono', 'Droid Sans Mono', 'Lucida Console', Consolas, Monaco, monospace; } h1 { font-size:2.2em; } h2 { font-size:1.8em; } h3 { font-size:1.4em; } h4 { font-size:1.0em; } h5 { font-size:0.9em; } h6 { font-size:0.8em; } a:visited { color: rgb(50%, 0%, 50%); } pre { margin-top: 0; max-width: 95%; border: 1px solid #ccc; white-space: pre-wrap; } pre code { display: block; padding: 0.5em; } code.r, code.cpp { background-color: #F8F8F8; } table, td, th { border: none; } blockquote { color:#666666; margin:0; padding-left: 1em; border-left: 0.5em #EEE solid; } hr { height: 0px; border-bottom: none; border-top-width: thin; border-top-style: dotted; border-top-color: #999999; } @media print { * { background: transparent !important; color: black !important; filter:none !important; -ms-filter: none !important; } body { font-size:12pt; max-width:100%; } a, a:visited { text-decoration: underline; } hr { visibility: hidden; page-break-before: always; } pre, blockquote { padding-right: 1em; page-break-inside: avoid; } tr, img { page-break-inside: avoid; } img { max-width: 100% !important; } @page :left { margin: 15mm 20mm 15mm 10mm; } @page :right { margin: 15mm 10mm 15mm 20mm; } p, h2, h3 { orphans: 3; widows: 3; } h2, h3 { page-break-after: avoid; } } pre .operator, pre .paren { color: rgb(104, 118, 135) } pre .literal { color: rgb(88, 72, 246) } pre .number { color: rgb(0, 0, 205); } pre .comment { color: rgb(76, 136, 107); } pre .keyword { color: rgb(0, 0, 255); } pre .identifier { color: rgb(0, 0, 0); } pre .string { color: rgb(3, 106, 7); } var hljs=new function(){function m(p){return p.replace(/&/gm,"&").replace(/"}while(y.length||w.length){var v=u().splice(0,1)[0];z+=m(x.substr(q,v.offset-q));q=v.offset;if(v.event=="start"){z+=t(v.node);s.push(v.node)}else{if(v.event=="stop"){var p,r=s.length;do{r--;p=s[r];z+=("")}while(p!=v.node);s.splice(r,1);while(r'+M[0]+""}else{r+=M[0]}O=P.lR.lastIndex;M=P.lR.exec(L)}return r+L.substr(O,L.length-O)}function J(L,M){if(M.sL&&e[M.sL]){var r=d(M.sL,L);x+=r.keyword_count;return r.value}else{return F(L,M)}}function I(M,r){var L=M.cN?'':"";if(M.rB){y+=L;M.buffer=""}else{if(M.eB){y+=m(r)+L;M.buffer=""}else{y+=L;M.buffer=r}}D.push(M);A+=M.r}function G(N,M,Q){var R=D[D.length-1];if(Q){y+=J(R.buffer+N,R);return false}var P=q(M,R);if(P){y+=J(R.buffer+N,R);I(P,M);return P.rB}var L=v(D.length-1,M);if(L){var O=R.cN?"":"";if(R.rE){y+=J(R.buffer+N,R)+O}else{if(R.eE){y+=J(R.buffer+N,R)+O+m(M)}else{y+=J(R.buffer+N+M,R)+O}}while(L1){O=D[D.length-2].cN?"":"";y+=O;L--;D.length--}var r=D[D.length-1];D.length--;D[D.length-1].buffer="";if(r.starts){I(r.starts,"")}return R.rE}if(w(M,R)){throw"Illegal"}}var E=e[B];var D=[E.dM];var A=0;var x=0;var y="";try{var s,u=0;E.dM.buffer="";do{s=p(C,u);var t=G(s[0],s[1],s[2]);u+=s[0].length;if(!t){u+=s[1].length}}while(!s[2]);if(D.length1){throw"Illegal"}return{r:A,keyword_count:x,value:y}}catch(H){if(H=="Illegal"){return{r:0,keyword_count:0,value:m(C)}}else{throw H}}}function g(t){var p={keyword_count:0,r:0,value:m(t)};var r=p;for(var q in e){if(!e.hasOwnProperty(q)){continue}var s=d(q,t);s.language=q;if(s.keyword_count+s.rr.keyword_count+r.r){r=s}if(s.keyword_count+s.rp.keyword_count+p.r){r=p;p=s}}if(r.language){p.second_best=r}return p}function i(r,q,p){if(q){r=r.replace(/^((]+|\t)+)/gm,function(t,w,v,u){return w.replace(/\t/g,q)})}if(p){r=r.replace(/\n/g,"")}return r}function n(t,w,r){var x=h(t,r);var v=a(t);var y,s;if(v){y=d(v,x)}else{return}var q=c(t);if(q.length){s=document.createElement("pre");s.innerHTML=y.value;y.value=k(q,c(s),x)}y.value=i(y.value,w,r);var u=t.className;if(!u.match("(\\s|^)(language-)?"+v+"(\\s|$)")){u=u?(u+" "+v):v}if(/MSIE [678]/.test(navigator.userAgent)&&t.tagName=="CODE"&&t.parentNode.tagName=="PRE"){s=t.parentNode;var p=document.createElement("div");p.innerHTML=""+y.value+"";t=p.firstChild.firstChild;p.firstChild.cN=s.cN;s.parentNode.replaceChild(p.firstChild,s)}else{t.innerHTML=y.value}t.className=u;t.result={language:v,kw:y.keyword_count,re:y.r};if(y.second_best){t.second_best={language:y.second_best.language,kw:y.second_best.keyword_count,re:y.second_best.r}}}function o(){if(o.called){return}o.called=true;var r=document.getElementsByTagName("pre");for(var p=0;p|=||=||=|\\?|\\[|\\{|\\(|\\^|\\^=|\\||\\|=|\\|\\||~";this.ER="(?![\\s\\S])";this.BE={b:"\\\\.",r:0};this.ASM={cN:"string",b:"'",e:"'",i:"\\n",c:[this.BE],r:0};this.QSM={cN:"string",b:'"',e:'"',i:"\\n",c:[this.BE],r:0};this.CLCM={cN:"comment",b:"//",e:"$"};this.CBLCLM={cN:"comment",b:"/\\*",e:"\\*/"};this.HCM={cN:"comment",b:"#",e:"$"};this.NM={cN:"number",b:this.NR,r:0};this.CNM={cN:"number",b:this.CNR,r:0};this.BNM={cN:"number",b:this.BNR,r:0};this.inherit=function(r,s){var p={};for(var q in r){p[q]=r[q]}if(s){for(var q in s){p[q]=s[q]}}return p}}();hljs.LANGUAGES.cpp=function(){var a={keyword:{"false":1,"int":1,"float":1,"while":1,"private":1,"char":1,"catch":1,"export":1,virtual:1,operator:2,sizeof:2,dynamic_cast:2,typedef:2,const_cast:2,"const":1,struct:1,"for":1,static_cast:2,union:1,namespace:1,unsigned:1,"long":1,"throw":1,"volatile":2,"static":1,"protected":1,bool:1,template:1,mutable:1,"if":1,"public":1,friend:2,"do":1,"return":1,"goto":1,auto:1,"void":2,"enum":1,"else":1,"break":1,"new":1,extern:1,using:1,"true":1,"class":1,asm:1,"case":1,typeid:1,"short":1,reinterpret_cast:2,"default":1,"double":1,register:1,explicit:1,signed:1,typename:1,"try":1,"this":1,"switch":1,"continue":1,wchar_t:1,inline:1,"delete":1,alignof:1,char16_t:1,char32_t:1,constexpr:1,decltype:1,noexcept:1,nullptr:1,static_assert:1,thread_local:1,restrict:1,_Bool:1,complex:1},built_in:{std:1,string:1,cin:1,cout:1,cerr:1,clog:1,stringstream:1,istringstream:1,ostringstream:1,auto_ptr:1,deque:1,list:1,queue:1,stack:1,vector:1,map:1,set:1,bitset:1,multiset:1,multimap:1,unordered_set:1,unordered_map:1,unordered_multiset:1,unordered_multimap:1,array:1,shared_ptr:1}};return{dM:{k:a,i:"",k:a,r:10,c:["self"]}]}}}();hljs.LANGUAGES.r={dM:{c:[hljs.HCM,{cN:"number",b:"\\b0[xX][0-9a-fA-F]+[Li]?\\b",e:hljs.IMMEDIATE_RE,r:0},{cN:"number",b:"\\b\\d+(?:[eE][+\\-]?\\d*)?L\\b",e:hljs.IMMEDIATE_RE,r:0},{cN:"number",b:"\\b\\d+\\.(?!\\d)(?:i\\b)?",e:hljs.IMMEDIATE_RE,r:1},{cN:"number",b:"\\b\\d+(?:\\.\\d*)?(?:[eE][+\\-]?\\d*)?i?\\b",e:hljs.IMMEDIATE_RE,r:0},{cN:"number",b:"\\.\\d+(?:[eE][+\\-]?\\d*)?i?\\b",e:hljs.IMMEDIATE_RE,r:1},{cN:"keyword",b:"(?:tryCatch|library|setGeneric|setGroupGeneric)\\b",e:hljs.IMMEDIATE_RE,r:10},{cN:"keyword",b:"\\.\\.\\.",e:hljs.IMMEDIATE_RE,r:10},{cN:"keyword",b:"\\.\\.\\d+(?![\\w.])",e:hljs.IMMEDIATE_RE,r:10},{cN:"keyword",b:"\\b(?:function)",e:hljs.IMMEDIATE_RE,r:2},{cN:"keyword",b:"(?:if|in|break|next|repeat|else|for|return|switch|while|try|stop|warning|require|attach|detach|source|setMethod|setClass)\\b",e:hljs.IMMEDIATE_RE,r:1},{cN:"literal",b:"(?:NA|NA_integer_|NA_real_|NA_character_|NA_complex_)\\b",e:hljs.IMMEDIATE_RE,r:10},{cN:"literal",b:"(?:NULL|TRUE|FALSE|T|F|Inf|NaN)\\b",e:hljs.IMMEDIATE_RE,r:1},{cN:"identifier",b:"[a-zA-Z.][a-zA-Z0-9._]*\\b",e:hljs.IMMEDIATE_RE,r:0},{cN:"operator",b:"|=|| Using R to Analyze G1GC Log Files Using R to Analyze G1GC Log Files Introduction Working in Oracle Platform Integration gives an engineer opportunities to work on a wide array of technologies. My team’s goal is to make Oracle applications run best on the Solaris/SPARC platform. When looking for bottlenecks in a modern applications, one needs to be aware of not only how the CPUs and operating system are executing, but also network, storage, and in some cases, the Java Virtual Machine. I was recently presented with about 1.5 GB of Java Garbage First Garbage Collector log file data. If you’re not familiar with the subject, you might want to review Garbage First Garbage Collector Tuning by Monica Beckwith. The customer had been running Java HotSpot 1.6.0_31 to host a web application server. I was told that the Solaris/SPARC server was running a Java process launched using a commmand line that included the following flags: -d64 -Xms9g -Xmx9g -XX:+UseG1GC -XX:MaxGCPauseMillis=200 -XX:InitiatingHeapOccupancyPercent=80 -XX:PermSize=256m -XX:MaxPermSize=256m -XX:+PrintGC -XX:+PrintGCTimeStamps -XX:+PrintHeapAtGC -XX:+PrintGCDateStamps -XX:+PrintFlagsFinal -XX:+DisableExplicitGC -XX:+UnlockExperimentalVMOptions -XX:ParallelGCThreads=8 Several sources on the internet indicate that if I were to print out the 1.5 GB of log files, it would require enough paper to fill the bed of a pick up truck. Of course, it would be fruitless to try to scan the log files by hand. Tools will be required to summarize the contents of the log files. Others have encountered large Java garbage collection log files. There are existing tools to analyze the log files: IBM’s GC toolkit The chewiebug GCViewer gchisto HPjmeter Instead of using one of the other tools listed, I decide to parse the log files with standard Unix tools, and analyze the data with R. Data Cleansing The log files arrived in two different formats. I guess that the difference is that one set of log files was generated using a more verbose option, maybe -XX:+PrintHeapAtGC, and the other set of log files was generated without that option. Format 1 In some of the log files, the log files with the less verbose format, a single trace, i.e. the report of a singe garbage collection event, looks like this: {Heap before GC invocations=12280 (full 61): garbage-first heap total 9437184K, used 7499918K [0xfffffffd00000000, 0xffffffff40000000, 0xffffffff40000000) region size 4096K, 1 young (4096K), 0 survivors (0K) compacting perm gen total 262144K, used 144077K [0xffffffff40000000, 0xffffffff50000000, 0xffffffff50000000) the space 262144K, 54% used [0xffffffff40000000, 0xffffffff48cb3758, 0xffffffff48cb3800, 0xffffffff50000000) No shared spaces configured. 2014-05-14T07:24:00.988-0700: 60586.353: [GC pause (young) 7324M->7320M(9216M), 0.1567265 secs] Heap after GC invocations=12281 (full 61): garbage-first heap total 9437184K, used 7496533K [0xfffffffd00000000, 0xffffffff40000000, 0xffffffff40000000) region size 4096K, 0 young (0K), 0 survivors (0K) compacting perm gen total 262144K, used 144077K [0xffffffff40000000, 0xffffffff50000000, 0xffffffff50000000) the space 262144K, 54% used [0xffffffff40000000, 0xffffffff48cb3758, 0xffffffff48cb3800, 0xffffffff50000000) No shared spaces configured. } A simple grep can be used to extract a summary: $ grep "\[ GC pause (young" g1gc.log 2014-05-13T13:24:35.091-0700: 3.109: [GC pause (young) 20M->5029K(9216M), 0.0146328 secs] 2014-05-13T13:24:35.440-0700: 3.459: [GC pause (young) 9125K->6077K(9216M), 0.0086723 secs] 2014-05-13T13:24:37.581-0700: 5.599: [GC pause (young) 25M->8470K(9216M), 0.0203820 secs] 2014-05-13T13:24:42.686-0700: 10.704: [GC pause (young) 44M->15M(9216M), 0.0288848 secs] 2014-05-13T13:24:48.941-0700: 16.958: [GC pause (young) 51M->20M(9216M), 0.0491244 secs] 2014-05-13T13:24:56.049-0700: 24.066: [GC pause (young) 92M->26M(9216M), 0.0525368 secs] 2014-05-13T13:25:34.368-0700: 62.383: [GC pause (young) 602M->68M(9216M), 0.1721173 secs] But that format wasn't easily read into R, so I needed to be a bit more tricky. I used the following Unix command to create a summary file that was easy for R to read. $ echo "SecondsSinceLaunch BeforeSize AfterSize TotalSize RealTime" $ grep "\[GC pause (young" g1gc.log | grep -v mark | sed -e 's/[A-SU-z,]/ /g' -e 's/->/ /' -e 's/: / /g' | more SecondsSinceLaunch BeforeSize AfterSize TotalSize RealTime 2014-05-13T13:24:35.091-0700 3.109 20 5029 9216 0.0146328 2014-05-13T13:24:35.440-0700 3.459 9125 6077 9216 0.0086723 2014-05-13T13:24:37.581-0700 5.599 25 8470 9216 0.0203820 2014-05-13T13:24:42.686-0700 10.704 44 15 9216 0.0288848 2014-05-13T13:24:48.941-0700 16.958 51 20 9216 0.0491244 2014-05-13T13:24:56.049-0700 24.066 92 26 9216 0.0525368 2014-05-13T13:25:34.368-0700 62.383 602 68 9216 0.1721173 Format 2 In some of the log files, the log files with the more verbose format, a single trace, i.e. the report of a singe garbage collection event, was more complicated than Format 1. Here is a text file with an example of a single G1GC trace in the second format. As you can see, it is quite complicated. It is nice that there is so much information available, but the level of detail can be overwhelming. I wrote this awk script (download) to summarize each trace on a single line. #!/usr/bin/env awk -f BEGIN { printf("SecondsSinceLaunch IncrementalCount FullCount UserTime SysTime RealTime BeforeSize AfterSize TotalSize\n") } ###################### # Save count data from lines that are at the start of each G1GC trace. # Each trace starts out like this: # {Heap before GC invocations=14 (full 0): # garbage-first heap total 9437184K, used 325496K [0xfffffffd00000000, 0xffffffff40000000, 0xffffffff40000000) ###################### /{Heap.*full/{ gsub ( "\\)" , "" ); nf=split($0,a,"="); split(a[2],b," "); getline; if ( match($0, "first") ) { G1GC=1; IncrementalCount=b[1]; FullCount=substr( b[3], 1, length(b[3])-1 ); } else { G1GC=0; } } ###################### # Pull out time stamps that are in lines with this format: # 2014-05-12T14:02:06.025-0700: 94.312: [GC pause (young), 0.08870154 secs] ###################### /GC pause/ { DateTime=$1; SecondsSinceLaunch=substr($2, 1, length($2)-1); } ###################### # Heap sizes are in lines that look like this: # [ 4842M->4838M(9216M)] ###################### /\[ .*]$/ { gsub ( "\\[" , "" ); gsub ( "\ \]" , "" ); gsub ( "->" , " " ); gsub ( "\$ " , " " ); gsub ( "\ $" , " " ); split($0,a," "); if ( split(a[1],b,"M") > 1 ) {BeforeSize=b[1]*1024;} if ( split(a[1],b,"K") > 1 ) {BeforeSize=b[1];} if ( split(a[2],b,"M") > 1 ) {AfterSize=b[1]*1024;} if ( split(a[2],b,"K") > 1 ) {AfterSize=b[1];} if ( split(a[3],b,"M") > 1 ) {TotalSize=b[1]*1024;} if ( split(a[3],b,"K") > 1 ) {TotalSize=b[1];} } ###################### # Emit an output line when you find input that looks like this: # [Times: user=1.41 sys=0.08, real=0.24 secs] ###################### /\[Times/ { if (G1GC==1) { gsub ( "," , "" ); split($2,a,"="); UserTime=a[2]; split($3,a,"="); SysTime=a[2]; split($4,a,"="); RealTime=a[2]; print DateTime,SecondsSinceLaunch,IncrementalCount,FullCount,UserTime,SysTime,RealTime,BeforeSize,AfterSize,TotalSize; G1GC=0; } } The resulting summary is about 25X smaller that the original file, but still difficult for a human to digest. SecondsSinceLaunch IncrementalCount FullCount UserTime SysTime RealTime BeforeSize AfterSize TotalSize ... 2014-05-12T18:36:34.669-0700: 3985.744 561 0 0.57 0.06 0.16 1724416 1720320 9437184 2014-05-12T18:36:34.839-0700: 3985.914 562 0 0.51 0.06 0.19 1724416 1720320 9437184 2014-05-12T18:36:35.069-0700: 3986.144 563 0 0.60 0.04 0.27 1724416 1721344 9437184 2014-05-12T18:36:35.354-0700: 3986.429 564 0 0.33 0.04 0.09 1725440 1722368 9437184 2014-05-12T18:36:35.545-0700: 3986.620 565 0 0.58 0.04 0.17 1726464 1722368 9437184 2014-05-12T18:36:35.726-0700: 3986.801 566 0 0.43 0.05 0.12 1726464 1722368 9437184 2014-05-12T18:36:35.856-0700: 3986.930 567 0 0.30 0.04 0.07 1726464 1723392 9437184 2014-05-12T18:36:35.947-0700: 3987.023 568 0 0.61 0.04 0.26 1727488 1723392 9437184 2014-05-12T18:36:36.228-0700: 3987.302 569 0 0.46 0.04 0.16 1731584 1724416 9437184 Reading the Data into R Once the GC log data had been cleansed, either by processing the first format with the shell script, or by processing the second format with the awk script, it was easy to read the data into R. g1gc.df = read.csv("summary.txt", row.names = NULL, stringsAsFactors=FALSE,sep="") str(g1gc.df) ## 'data.frame': 8307 obs. of 10 variables: ## $ row.names : chr "2014-05-12T14:00:32.868-0700:" "2014-05-12T14:00:33.179-0700:" "2014-05-12T14:00:33.677-0700:" "2014-05-12T14:00:35.538-0700:" ... ## $ SecondsSinceLaunch: num 1.16 1.47 1.97 3.83 6.1 ... ## $ IncrementalCount : int 0 1 2 3 4 5 6 7 8 9 ... ## $ FullCount : int 0 0 0 0 0 0 0 0 0 0 ... ## $ UserTime : num 0.11 0.05 0.04 0.21 0.08 0.26 0.31 0.33 0.34 0.56 ... ## $ SysTime : num 0.04 0.01 0.01 0.05 0.01 0.06 0.07 0.06 0.07 0.09 ... ## $ RealTime : num 0.02 0.02 0.01 0.04 0.02 0.04 0.05 0.04 0.04 0.06 ... ## $ BeforeSize : int 8192 5496 5768 22528 24576 43008 34816 53248 55296 93184 ... ## $ AfterSize : int 1400 1672 2557 4907 7072 14336 16384 18432 19456 21504 ... ## $ TotalSize : int 9437184 9437184 9437184 9437184 9437184 9437184 9437184 9437184 9437184 9437184 ... head(g1gc.df) ## row.names SecondsSinceLaunch IncrementalCount ## 1 2014-05-12T14:00:32.868-0700: 1.161 0 ## 2 2014-05-12T14:00:33.179-0700: 1.472 1 ## 3 2014-05-12T14:00:33.677-0700: 1.969 2 ## 4 2014-05-12T14:00:35.538-0700: 3.830 3 ## 5 2014-05-12T14:00:37.811-0700: 6.103 4 ## 6 2014-05-12T14:00:41.428-0700: 9.720 5 ## FullCount UserTime SysTime RealTime BeforeSize AfterSize TotalSize ## 1 0 0.11 0.04 0.02 8192 1400 9437184 ## 2 0 0.05 0.01 0.02 5496 1672 9437184 ## 3 0 0.04 0.01 0.01 5768 2557 9437184 ## 4 0 0.21 0.05 0.04 22528 4907 9437184 ## 5 0 0.08 0.01 0.02 24576 7072 9437184 ## 6 0 0.26 0.06 0.04 43008 14336 9437184 Basic Statistics Once the data has been read into R, simple statistics are very easy to generate. All of the numbers from high school statistics are available via simple commands. For example, generate a summary of every column: summary(g1gc.df) ## row.names SecondsSinceLaunch IncrementalCount FullCount ## Length:8307 Min. : 1 Min. : 0 Min. : 0.0 ## Class :character 1st Qu.: 9977 1st Qu.:2048 1st Qu.: 0.0 ## Mode :character Median :12855 Median :4136 Median : 12.0 ## Mean :12527 Mean :4156 Mean : 31.6 ## 3rd Qu.:15758 3rd Qu.:6262 3rd Qu.: 61.0 ## Max. :55484 Max. :8391 Max. :113.0 ## UserTime SysTime RealTime BeforeSize ## Min. :0.040 Min. :0.0000 Min. : 0.0 Min. : 5476 ## 1st Qu.:0.470 1st Qu.:0.0300 1st Qu.: 0.1 1st Qu.:5137920 ## Median :0.620 Median :0.0300 Median : 0.1 Median :6574080 ## Mean :0.751 Mean :0.0355 Mean : 0.3 Mean :5841855 ## 3rd Qu.:0.920 3rd Qu.:0.0400 3rd Qu.: 0.2 3rd Qu.:7084032 ## Max. :3.370 Max. :1.5600 Max. :488.1 Max. :8696832 ## AfterSize TotalSize ## Min. : 1380 Min. :9437184 ## 1st Qu.:5002752 1st Qu.:9437184 ## Median :6559744 Median :9437184 ## Mean :5785454 Mean :9437184 ## 3rd Qu.:7054336 3rd Qu.:9437184 ## Max. :8482816 Max. :9437184 Q: What is the total amount of User CPU time spent in garbage collection? sum(g1gc.df$UserTime) ## [1] 6236 As you can see, less than two hours of CPU time was spent in garbage collection. Is that too much? To find the percentage of time spent in garbage collection, divide the number above by total_elapsed_time*CPU_count. In this case, there are a lot of CPU’s and it turns out the the overall amount of CPU time spent in garbage collection isn’t a problem when viewed in isolation. When calculating rates, i.e. events per unit time, you need to ask yourself if the rate is homogenous across the time period in the log file. Does the log file include spikes of high activity that should be separately analyzed? Averaging in data from nights and weekends with data from business hours may alias problems. If you have a reason to suspect that the garbage collection rates include peaks and valleys that need independent analysis, see the “Time Series” section, below. Q: How much garbage is collected on each pass? The amount of heap space that is recovered per GC pass is surprisingly low: At least one collection didn’t recover any data. (“Min.=0”) 25% of the passes recovered 3MB or less. (“1st Qu.=3072”) Half of the GC passes recovered 4MB or less. (“Median=4096”) The average amount recovered was 56MB. (“Mean=56390”) 75% of the passes recovered 36MB or less. (“3rd Qu.=36860”) At least one pass recovered 2GB. (“Max.=2121000”) g1gc.df$Delta = g1gc.df$BeforeSize - g1gc.df$AfterSize summary(g1gc.df$Delta) ## Min. 1st Qu. Median Mean 3rd Qu. Max. ## 0 3070 4100 56400 36900 2120000 Q: What is the maximum User CPU time for a single collection? The worst garbage collection (“Max.”) is many standard deviations away from the mean. The data appears to be right skewed. summary(g1gc.df$UserTime) ## Min. 1st Qu. Median Mean 3rd Qu. Max. ## 0.040 0.470 0.620 0.751 0.920 3.370 sd(g1gc.df$UserTime) ## [1] 0.3966 Basic Graphics Once the data is in R, it is trivial to plot the data with formats including dot plots, line charts, bar charts (simple, stacked, grouped), pie charts, boxplots, scatter plots histograms, and kernel density plots. Histogram of User CPU Time per Collection I don't think that this graph requires any explanation. hist(g1gc.df$UserTime, main="User CPU Time per Collection", xlab="Seconds", ylab="Frequency") Box plot to identify outliers When the initial data is viewed with a box plot, you can see the one crazy outlier in the real time per GC. Save this data point for future analysis and drop the outlier so that it’s not throwing off our statistics. Now the box plot shows many outliers, which will be examined later, using times series analysis. Notice that the scale of the x-axis changes drastically once the crazy outlier is removed. par(mfrow=c(2,1)) boxplot(g1gc.df$UserTime,g1gc.df$SysTime,g1gc.df$RealTime, main="Box Plot of Time per GC\n(dominated by a crazy outlier)", names=c("usr","sys","elapsed"), xlab="Seconds per GC", ylab="Time (Seconds)", horizontal = TRUE, outcol="red") crazy.outlier.df=g1gc.df[g1gc.df$RealTime > 400,] g1gc.df=g1gc.df[g1gc.df$RealTime < 400,] boxplot(g1gc.df$UserTime,g1gc.df$SysTime,g1gc.df$RealTime, main="Box Plot of Time per GC\n(crazy outlier excluded)", names=c("usr","sys","elapsed"), xlab="Seconds per GC", ylab="Time (Seconds)", horizontal = TRUE, outcol="red") box(which = "outer", lty = "solid") Here is the crazy outlier for future analysis: crazy.outlier.df ## row.names SecondsSinceLaunch IncrementalCount ## 8233 2014-05-12T23:15:43.903-0700: 20741 8316 ## FullCount UserTime SysTime RealTime BeforeSize AfterSize TotalSize ## 8233 112 0.55 0.42 488.1 8381440 8235008 9437184 ## Delta ## 8233 146432 R Time Series Data To analyze the garbage collection as a time series, I’ll use Z’s Ordered Observations (zoo). “zoo is the creator for an S3 class of indexed totally ordered observations which includes irregular time series.” require(zoo) ## Loading required package: zoo ## ## Attaching package: 'zoo' ## ## The following objects are masked from 'package:base': ## ## as.Date, as.Date.numeric head(g1gc.df[,1]) ## [1] "2014-05-12T14:00:32.868-0700:" "2014-05-12T14:00:33.179-0700:" ## [3] "2014-05-12T14:00:33.677-0700:" "2014-05-12T14:00:35.538-0700:" ## [5] "2014-05-12T14:00:37.811-0700:" "2014-05-12T14:00:41.428-0700:" options("digits.secs"=3) times=as.POSIXct( g1gc.df[,1], format="%Y-%m-%dT%H:%M:%OS%z:") g1gc.z = zoo(g1gc.df[,-c(1)], order.by=times) head(g1gc.z) ## SecondsSinceLaunch IncrementalCount FullCount ## 2014-05-12 17:00:32.868 1.161 0 0 ## 2014-05-12 17:00:33.178 1.472 1 0 ## 2014-05-12 17:00:33.677 1.969 2 0 ## 2014-05-12 17:00:35.538 3.830 3 0 ## 2014-05-12 17:00:37.811 6.103 4 0 ## 2014-05-12 17:00:41.427 9.720 5 0 ## UserTime SysTime RealTime BeforeSize AfterSize ## 2014-05-12 17:00:32.868 0.11 0.04 0.02 8192 1400 ## 2014-05-12 17:00:33.178 0.05 0.01 0.02 5496 1672 ## 2014-05-12 17:00:33.677 0.04 0.01 0.01 5768 2557 ## 2014-05-12 17:00:35.538 0.21 0.05 0.04 22528 4907 ## 2014-05-12 17:00:37.811 0.08 0.01 0.02 24576 7072 ## 2014-05-12 17:00:41.427 0.26 0.06 0.04 43008 14336 ## TotalSize Delta ## 2014-05-12 17:00:32.868 9437184 6792 ## 2014-05-12 17:00:33.178 9437184 3824 ## 2014-05-12 17:00:33.677 9437184 3211 ## 2014-05-12 17:00:35.538 9437184 17621 ## 2014-05-12 17:00:37.811 9437184 17504 ## 2014-05-12 17:00:41.427 9437184 28672 Example of Two Benchmark Runs in One Log File The data in the following graph is from a different log file, not the one of primary interest to this article. I’m including this image because it is an example of idle periods followed by busy periods. It would be uninteresting to average the rate of garbage collection over the entire log file period. More interesting would be the rate of garbage collect in the two busy periods. Are they the same or different? Your production data may be similar, for example, bursts when employees return from lunch and idle times on weekend evenings, etc. Once the data is in an R Time Series, you can analyze isolated time windows. Clipping the Time Series data Flashing back to our test case… Viewing the data as a time series is interesting. You can see that the work intensive time period is between 9:00 PM and 3:00 AM. Lets clip the data to the interesting period: par(mfrow=c(2,1)) plot(g1gc.z$UserTime, type="h", main="User Time per GC\nTime: Complete Log File", xlab="Time of Day", ylab="CPU Seconds per GC", col="#1b9e77") clipped.g1gc.z=window(g1gc.z, start=as.POSIXct("2014-05-12 21:00:00"), end=as.POSIXct("2014-05-13 03:00:00")) plot(clipped.g1gc.z$UserTime, type="h", main="User Time per GC\nTime: Limited to Benchmark Execution", xlab="Time of Day", ylab="CPU Seconds per GC", col="#1b9e77") box(which = "outer", lty = "solid") Cumulative Incremental and Full GC count Here is the cumulative incremental and full GC count. When the line is very steep, it indicates that the GCs are repeating very quickly. Notice that the scale on the Y axis is different for full vs. incremental. plot(clipped.g1gc.z[,c(2:3)], main="Cumulative Incremental and Full GC count", xlab="Time of Day", col="#1b9e77") GC Analysis of Benchmark Execution using Time Series data In the following series of 3 graphs: The “After Size” show the amount of heap space in use after each garbage collection. Many Java objects are still referenced, i.e. alive, during each garbage collection. This may indicate that the application has a memory leak, or may indicate that the application has a very large memory footprint. Typically, an application's memory footprint plateau's in the early stage of execution. One would expect this graph to have a flat top. The steep decline in the heap space may indicate that the application crashed after 2:00. The second graph shows that the outliers in real execution time, discussed above, occur near 2:00. when the Java heap seems to be quite full. The third graph shows that Full GCs are infrequent during the first few hours of execution. The rate of Full GC's, (the slope of the cummulative Full GC line), changes near midnight. plot(clipped.g1gc.z[,c("AfterSize","RealTime","FullCount")], xlab="Time of Day", col=c("#1b9e77","red","#1b9e77")) GC Analysis of heap recovered Each GC trace includes the amount of heap space in use before and after the individual GC event. During garbage coolection, unreferenced objects are identified, the space holding the unreferenced objects is freed, and thus, the difference in before and after usage indicates how much space has been freed. The following box plot and bar chart both demonstrate the same point - the amount of heap space freed per garbage colloection is surprisingly low. par(mfrow=c(2,1)) boxplot(as.vector(clipped.g1gc.z$Delta), main="Amount of Heap Recovered per GC Pass", xlab="Size in KB", horizontal = TRUE, col="red") hist(as.vector(clipped.g1gc.z$Delta), main="Amount of Heap Recovered per GC Pass", xlab="Size in KB", breaks=100, col="red") box(which = "outer", lty = "solid") This graph is the most interesting. The dark blue area shows how much heap is occupied by referenced Java objects. This represents memory that holds live data. The red fringe at the top shows how much data was recovered after each garbage collection. barplot(clipped.g1gc.z[,c("AfterSize","Delta")], col=c("#7570b3","#e7298a"), xlab="Time of Day", border=NA) legend("topleft", c("Live Objects","Heap Recovered on GC"), fill=c("#7570b3","#e7298a")) box(which = "outer", lty = "solid") When I discuss the data in the log files with the customer, I will ask for an explaination for the large amount of referenced data resident in the Java heap. There are two are posibilities: There is a memory leak and the amount of space required to hold referenced objects will continue to grow, limited only by the maximum heap size. After the maximum heap size is reached, the JVM will throw an “Out of Memory” exception every time that the application tries to allocate a new object. If this is the case, the aplication needs to be debugged to identify why old objects are referenced when they are no longer needed. The application has a legitimate requirement to keep a large amount of data in memory. The customer may want to further increase the maximum heap size. Another possible solution would be to partition the application across multiple cluster nodes, where each node has responsibility for managing a unique subset of the data. Conclusion In conclusion, R is a very powerful tool for the analysis of Java garbage collection log files. The primary difficulty is data cleansing so that information can be read into an R data frame. Once the data has been read into R, a rich set of tools may be used for thorough evaluation.

Search Results

Search found 13249 results on 530 pages for 'performance tuning'.

Page 498/530 | < Previous Page | 494 495 496 497 498 499 500 501 502 503 504 505 | Next Page >

- by A Team Lead

- by Roman Schindlauer

- by Janice J. Heiss

- by jamiet

- by frank.nimphius

- by George Mamaladze

- by jmorourke

- by cibrax

- by Tanu Sood

- by BuckWoody

- by kellsey.ruppel

- by James Michael Hare

- by jsavit

- by Ashok_Ora

- by Joe Mayo

- by Oracle Accelerate for Midsize Companies

- by Maureen Byrne

- by Tanu Sood

- by Charles Lamb

- by Sylvie MacKenzie, PMP

- by FelixWehmeyer

- by drsql

- by usha

- by agallego

- by user12620111

< Previous Page | 494 495 496 497 498 499 500 501 502 503 504 505 | Next Page >