Search Results

Search found 10115 results on 405 pages for 'coding practices'.

Page 36/405 | < Previous Page | 32 33 34 35 36 37 38 39 40 41 42 43  | Next Page >

  • Congratulations 2010 Microsoft MVP, woo hoo 5th time now

    - by ssqa.net
    Well, its April 01st again and a big day for me 2 important events to note (daughter's birthday and MVP renewal notification). After a long travel from London to Hyderabad after speaking at UK AIC 2010 conference, I was able to make it by half day here for my daughter's birthday, phew. Then next one awaiting official confirmation about MVP renewal (April - Mar cycle), woo hooo here is one.... Dear Satya Jayanty, Congratulations! We are pleased to present you with the 2010 Microsoft® MVP Award...(read more)

    Read the article

  • Partition Wise Joins II

    - by jean-pierre.dijcks
    One of the things that I did not talk about in the initial partition wise join post was the effect it has on resource allocation on the database server. When Oracle applies a different join method - e.g. not PWJ - what you will see in SQL Monitor (in Enterprise Manager) or in an Explain Plan is a set of producers and a set of consumers. The producers scan the tables in the the join. If there are two tables the producers first scan one table, then the other. The producers thus provide data to the consumers, and when the consumers have the data from both scans they do the join and give the data to the query coordinator. Now that behavior means that if you choose a degree of parallelism of 4 to run such query with, Oracle will allocate 8 parallel processes. Of these 8 processes 4 are producers and 4 are consumers. The consumers only actually do work once the producers are fully done with scanning both sides of the join. In the plan above you can see that the producers access table SALES [line 11] and then do a PX SEND [line 9]. That is the producer set of processes working. The consumers receive that data [line 8] and twiddle their thumbs while the producers go on and scan CUSTOMERS. The producers send that data to the consumer indicated by PX SEND [line 5]. After receiving that data [line 4] the consumers do the actual join [line 3] and give the data to the QC [line 2]. BTW, the myth that you see twice the number of processes due to the setting PARALLEL_THREADS_PER_CPU=2 is obviously not true. The above is why you will see 2 times the processes of the DOP. In a PWJ plan the consumers are not present. Instead of producing rows and giving those to different processes, a PWJ only uses a single set of processes. Each process reads its piece of the join across the two tables and performs the join. The plan here is notably different from the initial plan. First of all the hash join is done right on top of both table scans [line 8]. This query is a little more complex than the previous so there is a bit of noise above that bit of info, but for this post, lets ignore that (sort stuff). The important piece here is that the PWJ plan typically will be faster and from a PX process number / resources typically cheaper. You may want to look out for those plans and try to get those to appear a lot... CREDITS: credits for the plans and some of the info on the plans go to Maria, as she actually produced these plans and is the expert on plans in general... You can see her talk about explaining the explain plan and other optimizer stuff over here: ODTUG in Washington DC, June 27 - July 1 On the Optimizer blog At OpenWorld in San Francisco, September 19 - 23 Happy joining and hope to see you all at ODTUG and OOW...

    Read the article

  • Avoiding new operator in JavaScript -- the better way

    - by greengit
    Warning: This is a long post. Let's keep it simple. I want to avoid having to prefix the new operator every time I call a constructor in JavaScript. This is because I tend to forget it, and my code screws up badly. The simple way around this is this... function Make(x) { if ( !(this instanceof arguments.callee) ) return new arguments.callee(x); // do your stuff... } But, I need this to accept variable no. of arguments, like this... m1 = Make(); m2 = Make(1,2,3); m3 = Make('apple', 'banana'); The first immediate solution seems to be the 'apply' method like this... function Make() { if ( !(this instanceof arguments.callee) ) return new arguments.callee.apply(null, arguments); // do your stuff } This is WRONG however -- the new object is passed to the apply method and NOT to our constructor arguments.callee. Now, I've come up with three solutions. My simple question is: which one seems best. Or, if you have a better method, tell it. First – use eval() to dynamically create JavaScript code that calls the constructor. function Make(/* ... */) { if ( !(this instanceof arguments.callee) ) { // collect all the arguments var arr = []; for ( var i = 0; arguments[i]; i++ ) arr.push( 'arguments[' + i + ']' ); // create code var code = 'new arguments.callee(' + arr.join(',') + ');'; // call it return eval( code ); } // do your stuff with variable arguments... } Second – Every object has __proto__ property which is a 'secret' link to its prototype object. Fortunately this property is writable. function Make(/* ... */) { var obj = {}; // do your stuff on 'obj' just like you'd do on 'this' // use the variable arguments here // now do the __proto__ magic // by 'mutating' obj to make it a different object obj.__proto__ = arguments.callee.prototype; // must return obj return obj; } Third – This is something similar to second solution. function Make(/* ... */) { // we'll set '_construct' outside var obj = new arguments.callee._construct(); // now do your stuff on 'obj' just like you'd do on 'this' // use the variable arguments here // you have to return obj return obj; } // now first set the _construct property to an empty function Make._construct = function() {}; // and then mutate the prototype of _construct Make._construct.prototype = Make.prototype; eval solution seems clumsy and comes with all the problems of "evil eval". __proto__ solution is non-standard and the "Great Browser of mIsERY" doesn't honor it. The third solution seems overly complicated. But with all the above three solutions, we can do something like this, that we can't otherwise... m1 = Make(); m2 = Make(1,2,3); m3 = Make('apple', 'banana'); m1 instanceof Make; // true m2 instanceof Make; // true m3 instanceof Make; // true Make.prototype.fire = function() { // ... }; m1.fire(); m2.fire(); m3.fire(); So effectively the above solutions give us "true" constructors that accept variable no. of arguments and don't require new. What's your take on this. -- UPDATE -- Some have said "just throw an error". My response is: we are doing a heavy app with 10+ constructors and I think it'd be far more wieldy if every constructor could "smartly" handle that mistake without throwing error messages on the console.

    Read the article

  • Programming Pearls (2nd Edition) vs More Programming Pearls: Confessions of a Coder [closed]

    - by Geek
    I have been reading very good reviews of the books by Jon Bentley : Programming Pearls (2nd Edition) More Programming Pearls: Confessions of a Coder. I know that these books have been out there for a long time and I feel bad that I haven't read either one . But it is always better late than never . I understand that the second one was written after the first one . So are these two books complementary to each other ? Do the second one assume that the reader has read the first one ? For some one who haven't read either which one would you propose to read up first ?

    Read the article

  • Progressive Enhancement vs. Single Page Apps

    - by SeanPlusPlus
    I just got back from a conference in Boston called An Event Apart. A really popular theme amongst the speakers was the idea of progressive enhancement - a site's content should go in the HTML, and JavaScript should only be used to enhance behavior. The arguments that the speakers gave for progressive enhancement were very compelling. Not only is it a solid pattern for supporting older browsers, and devices on a network with low bandwidth, but HTML fails much more gracefully than JavaScript (i.e. markup that is not supported is just ignored, while if a browser throws an exception while executing your script - you are hosed). Jeremy Keith gave a particularly insightful talk about this. But what about single page web apps like Backbone and Angular? The whole design behind these frameworks seems to push the developer toward moving content out of the HTML, and into something like a JSON API. I can not seem to gel these two design patterns: progressive enhancement vs. single page web apps. Are there instances when one is better than the other? Or are they not even antagonistic technologies, and I am missing something here with my mental model?

    Read the article

  • Multiple classes in a single .cs file - good or bad?

    - by Sergio
    Is it advisable to create multiple classes within a .cs file or should each .cs file have an individual class? For example: public class Items { public class Animal { } public class Person { } public class Object { } } Dodging the fact for a minute that this is a poor example of good architecture, is having more than a single class in a .cs file a code smell?

    Read the article

  • Running SSIS packages from C#

    - by Piotr Rodak
    Most of the developers and DBAs know about two ways of deploying packages: You can deploy them to database server and run them using SQL Server Agent job or you can deploy the packages to file system and run them using dtexec.exe utility. Both approaches have their pros and cons. However I would like to show you that there is a third way (sort of) that is often overlooked, and it can give you capabilities the ‘traditional’ approaches can’t. I have been working for a few years with applications that run packages from host applications that are implemented in .NET. As you know, SSIS provides programming model that you can use to implement more flexible solutions. SSIS applications are usually thought to be batch oriented, with fairly rigid architecture and processing model, with fixed timeframes when the packages are executed to process data. It doesn’t to be the case, you don’t have to limit yourself to batch oriented architecture. I have very good experiences with service oriented architectures processing large amounts of data. These applications are more complex than what I would like to show here, but the principle stays the same: you can execute packages as a service, on ad-hoc basis. You can also implement and schedule various signals, HTTP calls, file drops, time schedules, Tibco messages and other to run the packages. You can implement event handler that will trigger execution of SSIS when a certain event occurs in StreamInsight stream. This post is just a small example of how you can use the API and other features to create a service that can run SSIS packages on demand. I thought it might be a good idea to implement a restful service that would listen to requests and execute appropriate actions. As it turns out, it is trivial in C#. The application is implemented as console application for the ease of debugging and running. In reality, you might want to implement the application as Windows service. To begin, you have to reference namespace System.ServiceModel.Web and then add a few lines of code: Uri baseAddress = new Uri("http://localhost:8011/");               WebServiceHost svcHost = new WebServiceHost(typeof(PackRunner), baseAddress);                           try             {                 svcHost.Open();                   Console.WriteLine("Service is running");                 Console.WriteLine("Press enter to stop the service.");                 Console.ReadLine();                   svcHost.Close();             }             catch (CommunicationException cex)             {                 Console.WriteLine("An exception occurred: {0}", cex.Message);                 svcHost.Abort();             } The interesting lines are 3, 7 and 13. In line 3 you create a WebServiceHost object. In line 7 you start listening on the defined URL and then in line 13 you shut down the service. As you have noticed, the WebServiceHost constructor is accepting type of an object (here: PackRunner) that will be instantiated as singleton and subsequently used to process the requests. This is the class where you put your logic, but to tell WebServiceHost how to use it, the class must implement an interface which declares methods to be used by the host. The interface itself must be ornamented with attribute ServiceContract. [ServiceContract]     public interface IPackRunner     {         [OperationContract]         [WebGet(UriTemplate = "runpack?package={name}")]         string RunPackage1(string name);           [OperationContract]         [WebGet(UriTemplate = "runpackwithparams?package={name}&rows={rows}")]         string RunPackage2(string name, int rows);     } Each method that is going to be used by WebServiceHost has to have attribute OperationContract, as well as WebGet or WebInvoke attribute. The detailed discussion of the available options is outside of scope of this post. I also recommend using more descriptive names to methods . Then, you have to provide the implementation of the interface: public class PackRunner : IPackRunner     {         ... There are two methods defined in this class. I think that since the full code is attached to the post, I will show only the more interesting method, the RunPackage2.   /// <summary> /// Runs package and sets some of its variables. /// </summary> /// <param name="name">Name of the package</param> /// <param name="rows">Number of rows to export</param> /// <returns></returns> public string RunPackage2(string name, int rows) {     try     {         string pkgLocation = ConfigurationManager.AppSettings["PackagePath"];           pkgLocation = Path.Combine(pkgLocation, name.Replace("\"", ""));           Console.WriteLine();         Console.WriteLine("Calling package {0} with parameter {1}.", name, rows);                  Application app = new Application();         Package pkg = app.LoadPackage(pkgLocation, null);           pkg.Variables["User::ExportRows"].Value = rows;         DTSExecResult pkgResults = pkg.Execute();         Console.WriteLine();         Console.WriteLine(pkgResults.ToString());         if (pkgResults == DTSExecResult.Failure)         {             Console.WriteLine();             Console.WriteLine("Errors occured during execution of the package:");             foreach (DtsError er in pkg.Errors)                 Console.WriteLine("{0}: {1}", er.ErrorCode, er.Description);             Console.WriteLine();             return "Errors occured during execution. Contact your support.";         }                  Console.WriteLine();         Console.WriteLine();         return "OK";     }     catch (Exception ex)     {         Console.WriteLine(ex);         return ex.ToString();     } }   The method accepts package name and number of rows to export. The packages are deployed to the file system. The path to the packages is configured in the application configuration file. This way, you can implement multiple services on the same machine, provided you also configure the URL for each instance appropriately. To run a package, you have to reference Microsoft.SqlServer.Dts.Runtime namespace. This namespace is implemented in Microsoft.SQLServer.ManagedDTS.dll which in my case was installed in the folder “C:\Program Files (x86)\Microsoft SQL Server\100\SDK\Assemblies”. Once you have done it, you can create an instance of Microsoft.SqlServer.Dts.Runtime.Application as in line 18 in the above snippet. It may be a good idea to create the Application object in the constructor of the PackRunner class, to avoid necessity of recreating it each time the service is invoked. Then, in line 19 you see that an instance of Microsoft.SqlServer.Dts.Runtime.Package is created. The method LoadPackage in its simplest form just takes package file name as the first parameter. Before you run the package, you can set its variables to certain values. This is a great way of configuring your packages without all the hassle with dtsConfig files. In the above code sample, variable “User:ExportRows” is set to value of the parameter “rows” of the method. Eventually, you execute the package. The method doesn’t throw exceptions, you have to test the result of execution yourself. If the execution wasn’t successful, you can examine collection of errors exposed by the package. These are the familiar errors you often see during development and debugging of the package. I you run the package from the code, you have opportunity to persist them or log them using your favourite logging framework. The package itself is very simple; it connects to my AdventureWorks database and saves number of rows specified in variable “User::ExportRows” to a file. You should know that before you run the package, you can change its connection strings, logging, events and many more. I attach solution with the test service, as well as a project with two test packages. To test the service, you have to run it and wait for the message saying that the host is started. Then, just type (or copy and paste) the below command to your browser. http://localhost:8011/runpackwithparams?package=%22ExportEmployees.dtsx%22&rows=12 When everything works fine, and you modified the package to point to your AdventureWorks database, you should see "OK” wrapped in xml: I stopped the database service to simulate invalid connection string situation. The output of the request is different now: And the service console window shows more information: As you see, implementing service oriented ETL framework is not a very difficult task. You have ability to configure the packages before you run them, you can implement logging that is consistent with the rest of your system. In application I have worked with we also have resource monitoring and execution control. We don’t allow to run more than certain number of packages to run simultaneously. This ensures we don’t strain the server and we use memory and CPUs efficiently. The attached zip file contains two projects. One is the package runner. It has to be executed with administrative privileges as it registers HTTP namespace. The other project contains two simple packages. This is really a cool thing, you should check it out!

    Read the article

  • OS Analytics - Deep Dive Into Your OS

    - by Eran_Steiner
    Enterprise Manager Ops Center provides a feature called "OS Analytics". This feature allows you to get a better understanding of how the Operating System is being utilized. You can research the historical usage as well as real time data. This post will show how you can benefit from OS Analytics and how it works behind the scenes. We will have a call to discuss this blog - please join us!Date: Thursday, November 1, 2012Time: 11:00 am, Eastern Daylight Time (New York, GMT-04:00)1. Go to https://oracleconferencing.webex.com/oracleconferencing/j.php?ED=209833067&UID=1512092402&PW=NY2JhMmFjMmFh&RT=MiMxMQ%3D%3D2. If requested, enter your name and email address.3. If a password is required, enter the meeting password: oracle1234. Click "Join". To join the teleconference:Call-in toll-free number:       1-866-682-4770  (US/Canada)      Other countries:                https://oracle.intercallonline.com/portlets/scheduling/viewNumbers/viewNumber.do?ownerNumber=5931260&audioType=RP&viewGa=true&ga=ONConference Code:       7629343#Security code:            7777# Here is quick summary of what you can do with OS Analytics in Ops Center: View historical charts and real time value of CPU, memory, network and disk utilization Find the top CPU and Memory processes in real time or at a certain historical day Determine proper monitoring thresholds based on historical data View Solaris services status details Drill down into a process details View the busiest zones if applicable Where to start To start with OS Analytics, choose the OS asset in the tree and click the Analytics tab. You can see the CPU utilization, Memory utilization and Network utilization, along with the current real time top 5 processes in each category (click the image to see a larger version):  In the above screen, you can click each of the top 5 processes to see a more detailed view of that process. Here is an example of one of the processes: One of the cool things is that you can see the process tree for this process along with some port binding and open file descriptors. On Solaris machines with zones, you get an extra level of tabs, allowing you to get more information on the different zones: This is a good way to see the busiest zones. For example, one zone may not take a lot of CPU but it can consume a lot of memory, or perhaps network bandwidth. To see the detailed Analytics for each of the zones, simply click each of the zones in the tree and go to its Analytics tab. Next, click the "Processes" tab to see real time information of all the processes on the machine: An interesting column is the "Target" column. If you configured Ops Center to work with Enterprise Manager Cloud Control, then the two products will talk to each other and Ops Center will display the correlated target from Cloud Control in this table. If you are only using Ops Center - this column will remain empty. Next, if you view a Solaris machine, you will have a "Services" tab: By default, all services will be displayed, but you can choose to display only certain states, for example, those in maintenance or the degraded ones. You can highlight a service and choose to view the details, where you can see the Dependencies, Dependents and also the location of the service log file (not shown in the picture as you need to scroll down to see the log file). The "Threshold" tab is particularly helpful - you can view historical trends of different monitored values and based on the graph - determine what the monitoring values should be: You can ask Ops Center to suggest monitoring levels based on the historical values or you can set your own. The different colors in the graph represent the current set levels: Red for critical, Yellow for warning and Blue for Information, allowing you to quickly see how they're positioned against real data. It's important to note that when looking at longer periods, Ops Center smooths out the data and uses averages. So when looking at values such as CPU Usage, try shorter time frames which are more detailed, such as one hour or one day. Applying new monitoring values When first applying new values to monitored attributes - a popup will come up asking if it's OK to get you out of the current Monitoring Policy. This is OK if you want to either have custom monitoring for a specific machine, or if you want to use this current machine as a "Gold image" and extract a Monitoring Policy from it. You can later apply the new Monitoring Policy to other machines and also set it as a default Monitoring Profile. Once you're done with applying the different monitoring values, you can review and change them in the "Monitoring" tab. You can also click the "Extract a Monitoring Policy" in the actions pane on the right to save all the new values to a new Monitoring Policy, which can then be found under "Plan Management" -> "Monitoring Policies". Visiting the past Under the "History" tab you can "go back in time". This is very helpful when you know that a machine was busy a few hours ago (perhaps in the middle of the night?), but you were not around to take a look at it in real time. Here's a view into yesterday's data on one of the machines: You can see an interesting CPU spike happening at around 3:30 am along with some memory use. In the bottom table you can see the top 5 CPU and Memory consumers at the requested time. Very quickly you can see that this spike is related to the Solaris 11 IPS repository synchronization process using the "pkgrecv" command. The "time machine" doesn't stop here - you can also view historical data to determine which of the zones was the busiest at a given time: Under the hood The data collected is stored on each of the agents under /var/opt/sun/xvm/analytics/historical/ An "os.zip" file exists for the main OS. Inside you will find many small text files, named after the Epoch time stamp in which they were taken If you have any zones, there will be a file called "guests.zip" containing the same small files for all the zones, as well as a folder with the name of the zone along with "os.zip" in it If this is the Enterprise Controller or the Proxy Controller, you will have folders called "proxy" and "sat" in which you will find the "os.zip" for that controller The actual script collecting the data can be viewed for debugging purposes as well: On Linux, the location is: /opt/sun/xvmoc/private/os_analytics/collect On Solaris, the location is /opt/SUNWxvmoc/private/os_analytics/collect If you would like to redirect all the standard error into a file for debugging, touch the following file and the output will go into it: # touch /tmp/.collect.stderr   The temporary data is collected under /var/opt/sun/xvm/analytics/.collectdb until it is zipped. If you would like to review the properties for the Analytics, you can view those per each agent in /opt/sun/n1gc/lib/XVM.properties. Find the section "Analytics configurable properties for OS and VSC" to view the Analytics specific values. I hope you find this helpful! Please post questions in the comments below. Eran Steiner

    Read the article

  • A new version of Oracle Enterprise Manager Ops Center Doctor (OCDoctor ) Utility released

    - by Anand Akela
    In February,  we posted a blog of Oracle Enterprise Manager Ops Center Doctor aka OCDoctor Utility. This utility assists in various stages of the Ops Center deployment and can be a real life saver. It is updated on a regular basis with additional knowledge (similar to an antivirus subscription) to help you identify and resolve known issues or suggest ways to improve performance.A new version ( Version 4.00 ) of the OCDoctor is now available . This new version adds full support for recently announced Oracle Enterprise Manager Ops Center 12c including prerequisites checks, troubleshoot tests, log collection, tuning and product metadata updates. In addition, it adds several bug fixes and enhancements to OCDoctor Utility.To download OCDoctor for new installations:https://updates.oracle.com/OCDoctor/OCDoctor-latest.zipFor existing installations, simply run:# /var/opt/sun/xvm/OCDoctor/OCDoctor.sh --updateTip : If you have Oracle Enterprise Manager Ops Center12c EC installed, your OCDoctor will automatically update overnight. Join Oracle Launch Webcast : Total Cloud Control for Systems on April 12th at 9 AM PST to learn more about  Oracle Enterprise Manager Ops Center 12c from Oracle Senior Vice President John Fowler, Oracle Vice President of Systems Management Steve Wilson and a panel of Oracle executive. Stay connected with  Oracle Enterprise Manager   :  Twitter | Facebook | YouTube | Linkedin | Newsletter

    Read the article

  • What happened to Alan Cooper's Unified File Model?

    - by PAUL Mansour
    For a long time Alan Cooper (in the 3 versions of his book "About Face") has been promoting a "unified file model" to, among other things, dispense with what he calls the most idiotic message box ever invented - the one the pops up when hit the close button on an app or form saying "Do you want to discard your changes?" I like the idea and his arguments, but also have the knee-jerk reaction against it that most seasoned programmers and users have. While Cooper's book seems quite popular and respected, there is remarkably little discussion of this particular issue on the Web that I can find. Petter Hesselberg, the author of "Programming Industrial Strength Windows" mentions it but that seems about it. I have an opportunity to implement this in the (desktop) project I am working on, but face resistance by customers and co-workers, who are of course familiar with the MS Word and Excel way of doing things. I'm in a position to override their objections, but am not sure if I should. My questions are: Are there any good discussions of this that I have failed to find? Is anyone doing this in their apps? Is it a good idea that it is unfortunately not practical to implement until, say, Microsoft does it?

    Read the article

  • Why do we keep using CSV?

    - by Stephen
    Why do we keep using CSV? I recently made a shift to working the health domain and despite the wonderful work in data transfer standards, all data transfer is in CSV, both for reporting to external organisations, and for data migrations when implementing new systems. Unfortunately the use of CSV is the cause of the endless repetition of the same stupid errors, with the same waste of developer time. (bad escaping, failing to handle null fields etc.) I know we can do better, and anything between JSON and XML (depending on the instance) would be fine. (Most of the time this is data going from one MS SQLserver 2005 to another!) I feel as if each time I see this happening I am literally watching one developer waste anothers time. So why do we keep shafting each other? When will we stop?

    Read the article

  • Best PHP-based web development 'stack' of 2011

    - by Jens Roland
    I have been building PHP-based web sites for many years, and lately it seems I'm discovering another interesting new tool or method once every few weeks. This begs the question - what is the current state of the art in PHP development stacks for the seasoned coder? I'm specifically interested in the following: High-performance web server Database MVC framework Build tool Revision control Continuous Integration Automated testing Non-persistent caching I'd like to optimize my stack for scalability and rapid development. I'm not looking for personal preference here, I'm looking for real, quantifiable reasons to pick this-over-that.

    Read the article

  • Feature pack for SQL Server 2005 SP4 - collection of standalone packages

    - by ssqa.net
    With the release of SQL2005Sp4 an additional task is essential for DBAs & Developers to avoid any compatibility issues with existing code agains SP4 instance. Feature pack for SQL Server 2005 SP4 is available to download which contains the standalone packages such as SQLNative Client, ADOMD, OLAPDM etc.... as it states the feature pack are built on latest versions of add-on and backward compatibility contents for SQL Server 2005. The above link provides individual file to download for each environment...(read more)

    Read the article

  • FairScheduling Conventions in Hadoop

    - by dan.mcclary
    While scheduling and resource allocation control has been present in Hadoop since 0.20, a lot of people haven't discovered or utilized it in their initial investigations of the Hadoop ecosystem. We could chalk this up to many things: Organizations are still determining what their dataflow and analysis workloads will comprise Small deployments under tests aren't likely to show the signs of strains that would send someone looking for resource allocation options The default scheduling options -- the FairScheduler and the CapacityScheduler -- are not placed in the most prominent position within the Hadoop documentation. However, for production deployments, it's wise to start with at least the foundations of scheduling in place so that you can tune the cluster as workloads emerge. To do that, we have to ask ourselves something about what the off-the-rack scheduling options are. We have some choices: The FairScheduler, which will work to ensure resource allocations are enforced on a per-job basis. The CapacityScheduler, which will ensure resource allocations are enforced on a per-queue basis. Writing your own implementation of the abstract class org.apache.hadoop.mapred.job.TaskScheduler is an option, but usually overkill. If you're going to have several concurrent users and leverage the more interactive aspects of the Hadoop environment (e.g. Pig and Hive scripting), the FairScheduler is definitely the way to go. In particular, we can do user-specific pools so that default users get their fair share, and specific users are given the resources their workloads require. To enable fair scheduling, we're going to need to do a couple of things. First, we need to tell the JobTracker that we want to use scheduling and where we're going to be defining our allocations. We do this by adding the following to the mapred-site.xml file in HADOOP_HOME/conf: <property> <name>mapred.jobtracker.taskScheduler</name> <value>org.apache.hadoop.mapred.FairScheduler</value> </property> <property> <name>mapred.fairscheduler.allocation.file</name> <value>/path/to/allocations.xml</value> </property> <property> <name>mapred.fairscheduler.poolnameproperty</name> <value>pool.name</value> </property> <property> <name>pool.name</name> <value>${user.name}</name> </property> What we've done here is simply tell the JobTracker that we'd like to task scheduling to use the FairScheduler class rather than a single FIFO queue. Moreover, we're going to be defining our resource pools and allocations in a file called allocations.xml For reference, the allocation file is read every 15s or so, which allows for tuning allocations without having to take down the JobTracker. Our allocation file is now going to look a little like this <?xml version="1.0"?> <allocations> <pool name="dan"> <minMaps>5</minMaps> <minReduces>5</minReduces> <maxMaps>25</maxMaps> <maxReduces>25</maxReduces> <minSharePreemptionTimeout>300</minSharePreemptionTimeout> </pool> <mapreduce.job.user.name="dan"> <maxRunningJobs>6</maxRunningJobs> </user> <userMaxJobsDefault>3</userMaxJobsDefault> <fairSharePreemptionTimeout>600</fairSharePreemptionTimeout> </allocations> In this case, I've explicitly set my username to have upper and lower bounds on the maps and reduces, and allotted myself double the number of running jobs. Now, if I run hive or pig jobs from either the console or via the Hue web interface, I'll be treated "fairly" by the JobTracker. There's a lot more tweaking that can be done to the allocations file, so it's best to dig down into the description and start trying out allocations that might fit your workload.

    Read the article

  • A Primer on Migrating Oracle Applications to a New Platform

    - by Nick Quarmby
    In Support we field a lot of questions about the migration of Oracle Applications to different platforms.  This article describes the techniques available for migrating an Oracle Applications environment to a new platform and discusses some of the common questions that arise during migration.  This subject has been frequently discussed in previous blog articles but there still seems to be a gap regarding the type of questions we are frequently asked in Service Requests. Some of the questions we see are quite abstract. Customers simply want to get a grip on understanding how they approach a migration. Others want to know if a particular architecture is viable. Other customers ask about mixing different platforms within a single Oracle Applications environment.    Just to clarify, throughout this article, the term 'platform' refers specifically to operating systems and not to the underlying hardware. For a clear definition of 'platform' in the context of Oracle Applications Support then Terri's very timely article:Oracle E-Business Suite Platform SmörgåsbordThe migration process is very similar for both 11i and R12 so this article only mentions specific differences where relevant.

    Read the article

  • Managing Personal Projects As Solo Developer - Getting out of depth and failing projects

    - by James Jeffery
    I need some advice on project management. I start a project, and often times it will a large project for a solo developer. Usually its a web project. I handle everything from the UI, to the JS, PHP, server management etc. Half way in I feel out of my depth. I lose where I am, so I spend a couple of days away from the project to avoid the stress and before you know it, it becomes another unfinished project. I try to use frameworks and code libraries to make my developments easier on myself. Sometimes I will complete a project so it "works" and then go back and handle errors, design the UI properly and stuff. But without fail I will always end up out of my depth. I've though about outsourcing tasks such as the UI, and the behaviour, and focusing just on the PHP - which I feel is my strong point. But then pride kicks in, and I don't feel at one with a project I haven't completed myself. Does this make sense? I am sure there are many others who have felt like this either at home, or at work, and I would love some advice on managing my projects better.

    Read the article

  • How do you KISS ?

    - by Conor
    The KISS principal is a highly quoted design mantra. The aim of this principle is to stamp out unnecessary complexity on a project. This is a good thing, saving time, energy and money. It can lead to a relatively stress free implementation and a simple, elegant, maintainable end product. A lot of discussion on KISS provides mechanisms to simplify requirements, design and implementation. Things that spring to mind include: avoid scope creep; simple obvious design and code; minimal run-time dependencies; refactoring to maintain simplicity; etc. However there are a lot of implicit things that we do to KISS. Examples: small team sizes; minimal management layers; tidy desk; mastery of a single IDE; clear concise error messages; scripts to automate/encapsulate tasks; etc The purpose of this question is to derive a checklist of KISS items. I'm especially interested in non-obvious items.

    Read the article

  • Tips/Process for web-development using Django in a small team

    - by Mridang Agarwalla
    We're developing a web app uing Django and we're a small team of 3-4 programmers — some doing the UI stuff and some doing the Backend stuff. I'd love some tips and suggestions from the people here. This is out current setup: We're using Git as as our SCM tool and following this branching model. We're following the PEP8 for your style guide. Agile is our software development methodology and we're using Jira for that. We're using the Confluence plugin for Jira for documentation and I'm going to be writing a script that also dumps the PyDocs into Confluence. We're using virtualenv for sandboxing We're using zc.buildout for building This is whatever I can think of off the top of my head. Any other suggestions/tips would be welcome. I feel that we have a pretty good set up but I'm also confident that we could do more. Thanks.

    Read the article

  • Managing user privileges, best practice.

    - by Loïc N.
    I'm am new to web development. I'm creating a website where different user can have different privileges, such as creating/editing/deleting a news, or adding/editing/deleting whatever kind of content on the website. I started by creating a "user type" that would indicate the user's privileges (such as "user", "newser", "moderator", "admin", and so on), but i quickly started noticing issues that made me think that this might be a naive approach to this issue. What if i want to give a regular user the right to edit a news (for whatever reason)? Then the user would be half "user", half "newser". But the system i use can only handle one user-type. So what would be the best practice here? I was thinking of removing the concept of roles (or "user-types" such as newser) and only have the concept of "privilege", where every user could have zero to many privileges. So, to re-use the above example, if i wanted a user to have the right to edit some news, i would only have to give him a "edit news" privilege. Is this the way to go?

    Read the article

  • How do you deal with translating theory into practice?

    - by Mr. Shickadance
    Hello all! Being a computer scientist in a research field I am often tasked with working alongside professionals outside of the software domain (think math people, electrical engineer etc), and then translating their theories and ideas into real-world implementations. I often find it difficult when they present a theoretical problem which appears to be somewhat disconnected from reality. I am not saying that the theory is bogus, only that it is difficult to translate into real-world situations. For example, recently I have been working with software defined radios. We are exploring many different areas, but often the math specialists in my group would present a problem which is heavily grounded in theory (signal processing, physics, whatever). I often struggle at times where it is hard to draw direct parallels between the theory and the real-world implementation that I need to develop. Say we are working on an energy detector, the theory person in my group would say "you need to measure the noise variance with no signal present." This leads me to think "how the hell do I isolate noise from a signal in reality?" There are many examples, but I hope you see where I am going. So, my question is how does one deal with implementation of theoretical concepts when the theory seems detached from reality? Or at least when the connections are not so clear. Or perhaps, the person with the 'theory' may be ignorant of real restrictions? Note: I found this to be a hard question to ask - hopefully you are following me. If you have suggestions on how I could improve it, by all means let me know! Thanks for looking! EDIT: To be a bit more clear, I understand in situations like this that I must learn that specific domain myself to an extent (i.e. signal processing), but I am more concerned with when those theoretical concepts do not appear to be as grounded in practice as one would like.

    Read the article

  • Keep Your Eye on the Ball

    - by [email protected]
    With the FIFA World Cup 2010 in South Africa almost a week underway, the soccer fans all around the World are talking about at least 2 things. That typical vuvuzela sound and the new Jabulani ball, saying it moves unpredictably, is difficult to handle and somehow the altitude of the World Cup stadiums also seem to be a contributing factor.(Picture taken from http://www.flickr.com/photos/warrenski/4143923059/ under a Creative Commons license)Although the FIFA states that it hasn't received any official complaints, the end users don't seem to be very happy with this new ball. This brings me to a comparison with IT management and testing. When you're in a situation where you're introducing a new product, in IT terms, introducing a new application, you would like to test all possible scenarios that your end users could be using and experiencing. However, that's a very time and resource intensive process to do for every application change or update.  It's like getting ready for the big game but you have no game plan.That's why a new approach has been developed. One that's based on the 80/20 rule. Testing 80% of the application will cost about 20% of the efforts. The remaining 20% of your application will not be tested before deployment, but monitored with a real user monitoring solution immediately after deployment. These tools track all user experiences, including error messages and the performance and availability metrics from an end user perspective. Should any anomaly occur, you would be able to repair it quickly so you and your end users can get back into the game.These real user sessions can be easily converted into testing scripts, so the 80% of the application testing can be complimented with the remaining 20%.Oracle Enterprise Manager 11g group of products offers both the real user monitoring solution with Oracle Real User Experience Insight, as well as the required testing solution with Oracle Application Testing Suite. Visit our Oracle Enterprise Manager 11g resource center and find out how it's Business-Driven IT Management approach will help you keep your eye on your business ball.Happy World Cup.

    Read the article

  • COLUMNS_UPDATED() for audit triggers

    - by Piotr Rodak
    In SQL Server 2005, triggers are pretty much the only option if you want to audit changes to a table. There are many ways you can decide to store the change information. You may decide to store every changed row as a whole, either in a history table or as xml in audit table. The former case requires having a history table with exactly same schema as the audited table, the latter makes data retrieval and management of the table a bit tricky. Both approaches also suffer from the tendency to consume...(read more)

    Read the article

  • How can I get my progress reviewed as a solo junior developer

    - by Oliver Hyde
    I am currently working for a 2 person company, as the solo primary developer. My boss gets the clients, mocks up some png design templates and hands them over to me. This system has been working fine and i'm really enjoying it. The types of projects I work on are for small - medium sized businesses and they usually want a CMS system. Developed from scratch i'll build a customised backend for the client to add/edit/remove categories, tags, products etc and then output them to the front end according to the design template handed to me. As time has gone on, the projects have increased in complexity, with shopping cart / ordering features and other common e-commerce type features. Again, this system has been working fine and i'm really enjoying it. My issue is my personal development as a programmer. I spend a lot of my spare time reading programming blogs, checking through stackexchange, reading suggested programming books (currently on 'The Pragmatic Programmer', really good so far), doing brain exercises (lumosity.com and khanacademy math problems), doing lots of physical exercise and other personal development type activities. I can't help but feel though, that I'm missing out on feedback, critique. My boss is great and never holds back on praise in regards to my work, but he unfortunately is either to busy to check my code, or to be honest, I don't think it's one of his specialties and so can't provide feedback. I want to know what i'm doing wrong and what i'm doing right. Should I be putting that much logic in the controller, am I modulating my code enough etc. So what I have done is developed a little 'Family Budgeting' app and tried to do it as cleanly and effectively as I currently know how. What i'm wanting to know is, is there somewhere I can submit this app, and have some seasoned developers provide feedback. It's not just a subsection of my code like 'codereview.stackexchange' appears to require, it's my entire workflow that I want critiqued. I know this is a lot to ask, and I expect the main advice given will be to look for a job within a team, which is certainly something I will look into later down the track, but for now I want to persist with my current employment situation, but just don't want to develop too many bad habits. Let me know if I can provide any further information to help clarify, or if this isn't the right place for this type of question I apologise in advance. Didn't want to use reddit as I felt this community fosters more well thought out responses.

    Read the article

  • Linked servers and performance impact: Direction matters!

    - by Linchi Shea
    When you have some data on a SQL Server instance (say SQL01) and you want to move the data to another SQL Server instance (say SQL02) through openquery(), you can either push the data from SQL01, or pull the data from SQL02. To push the data, you can run a SQL script like the following on SQL01, which is the source server: -- The push script -- Run this on SQL01 use testDB go insert openquery(SQL02, 'select * from testDB.dbo.target_table') select * from source_table; To pull the data, you can run...(read more)

    Read the article

  • Database design and performance impact

    - by Craige
    I have a database design issue that I'm not quite sure how to approach, nor if the benefits out weigh the costs. I'm hoping some P.SE members can give some feedback on my suggested design, as well as any similar experiences they may have came across. As it goes, I am building an application that has large reporting demands. Speed is an important issue, as there will be peak usages throughout the year. This application/database has a multiple-level, many-to-many relationship. eg object a object b object c object d object b has relationship to object a object c has relationship to object b, a object d has relationship to object c, b, a Theoretically, this could go on for unlimited levels, though logic dictates it could only go so far. My idea here, to speed up reporting, would be to create a syndicate table that acts as a global many-to-many join table. In this table (with the given example), one might see: +----------+-----------+---------+ | child_id | parent_id | type_id | +----------+-----------+---------+ | b | a | 1 | | c | b | 2 | | c | a | 3 | | d | c | 4 | | d | b | 5 | | d | a | 6 | +----------+-----------+---------+ Where a, b, c and d would translate to their respective ID's in their respective tables. So, for ease of reporting all of a which exist on object d, one could query SELECT * FROM `syndicates` ... JOINS TO child and parent tables ... WHERE parent_id=a and type_id=6; rather than having a query with a join to each level up the chain. The Problem This table grows exponentially, and in a given year, could easily grow past 20,000 records for one client. Given multiple clients over multiple years, this table will VERY quickly explode to millions of records and beyond. Now, the database will, in time, be partitioned across multiple servers, but I would like (as most would) to keep the number of servers as low as possible while still offering flexibility. Also writes and updates would be exponentially longer (though possibly not noticeable to the end user) as there would be multiple inserts/updates/scans on this table to keep it in sync. Am I going in the right direction here, or am I way off track. What would you do in a similar situation? This solution seems overly complex, but allows the greatest flexibility and fastest read-operations. Sidenote 1 - This structure allows me to add new levels to the tree easily. Sidenote 2 - The database querying for this database is done through an ORM framework.

    Read the article

< Previous Page | 32 33 34 35 36 37 38 39 40 41 42 43  | Next Page >