Search Results

Search found 1993 results on 80 pages for 'comparison'.

Page 72/80 | < Previous Page | 68 69 70 71 72 73 74 75 76 77 78 79  | Next Page >

• Big Data – Role of Cloud Computing in Big Data – Day 11 of 21

  - by Pinal Dave

  In yesterday’s blog post we learned the importance of the NewSQL. In this article we will understand the role of Cloud in Big Data Story What is Cloud? Cloud is the biggest buzzword around from last few years. Everyone knows about the Cloud and it is extremely well defined online. In this article we will discuss cloud in the context of the Big Data. Cloud computing is a method of providing a shared computing resources to the application which requires dynamic resources. These resources include applications, computing, storage, networking, development and various deployment platforms. The fundamentals of the cloud computing are that it shares pretty much share all the resources and deliver to end users as a service.  Examples of the Cloud Computing and Big Data are Google and Amazon.com. Both have fantastic Big Data offering with the help of the cloud. We will discuss this later in this blog post. There are two different Cloud Deployment Models: 1) The Public Cloud and 2) The Private Cloud Public Cloud Public Cloud is the cloud infrastructure build by commercial providers (Amazon, Rackspace etc.) creates a highly scalable data center that hides the complex infrastructure from the consumer and provides various services. Private Cloud Private Cloud is the cloud infrastructure build by a single organization where they are managing highly scalable data center internally. Here is the quick comparison between Public Cloud and Private Cloud from Wikipedia:   Public Cloud Private Cloud Initial cost Typically zero Typically high Running cost Unpredictable Unpredictable Customization Impossible Possible Privacy No (Host has access to the data Yes Single sign-on Impossible Possible Scaling up Easy while within defined limits Laborious but no limits Hybrid Cloud Hybrid Cloud is the cloud infrastructure build with the composition of two or more clouds like public and private cloud. Hybrid cloud gives best of the both the world as it combines multiple cloud deployment models together. Cloud and Big Data – Common Characteristics There are many characteristics of the Cloud Architecture and Cloud Computing which are also essentially important for Big Data as well. They highly overlap and at many places it just makes sense to use the power of both the architecture and build a highly scalable framework. Here is the list of all the characteristics of cloud computing important in Big Data Scalability Elasticity Ad-hoc Resource Pooling Low Cost to Setup Infastructure Pay on Use or Pay as you Go Highly Available Leading Big Data Cloud Providers There are many players in Big Data Cloud but we will list a few of the known players in this list. Amazon Amazon is arguably the most popular Infrastructure as a Service (IaaS) provider. The history of how Amazon started in this business is very interesting. They started out with a massive infrastructure to support their own business. Gradually they figured out that their own resources are underutilized most of the time. They decided to get the maximum out of the resources they have and hence  they launched their Amazon Elastic Compute Cloud (Amazon EC2) service in 2006. Their products have evolved a lot recently and now it is one of their primary business besides their retail selling. Amazon also offers Big Data services understand Amazon Web Services. Here is the list of the included services: Amazon Elastic MapReduce – It processes very high volumes of data Amazon DynammoDB – It is fully managed NoSQL (Not Only SQL) database service Amazon Simple Storage Services (S3) – A web-scale service designed to store and accommodate any amount of data Amazon High Performance Computing – It provides low-tenancy tuned high performance computing cluster Amazon RedShift – It is petabyte scale data warehousing service Google Though Google is known for Search Engine, we all know that it is much more than that. Google Compute Engine – It offers secure, flexible computing from energy efficient data centers Google Big Query – It allows SQL-like queries to run against large datasets Google Prediction API – It is a cloud based machine learning tool Other Players Besides Amazon and Google we also have other players in the Big Data market as well. Microsoft is also attempting Big Data with the Cloud with Microsoft Azure. Additionally Rackspace and NASA together have initiated OpenStack. The goal of Openstack is to provide a massively scaled, multitenant cloud that can run on any hardware. Thing to Watch The cloud based solutions provides a great integration with the Big Data’s story as well it is very economical to implement as well. However, there are few things one should be very careful when deploying Big Data on cloud solutions. Here is a list of a few things to watch: Data Integrity Initial Cost Recurring Cost Performance Data Access Security Location Compliance Every company have different approaches to Big Data and have different rules and regulations. Based on various factors, one can implement their own custom Big Data solution on a cloud. Tomorrow In tomorrow’s blog post we will discuss about various Operational Databases supporting Big Data. Reference: Pinal Dave (http://blog.sqlauthority.com) Filed under: Big Data, PostADay, SQL, SQL Authority, SQL Query, SQL Server, SQL Tips and Tricks, T SQL

  Read the article

• Why lock-free data structures just aren't lock-free enough

  - by Alex.Davies

  Today's post will explore why the current ways to communicate between threads don't scale, and show you a possible way to build scalable parallel programming on top of shared memory. The problem with shared memory Soon, we will have dozens, hundreds and then millions of cores in our computers. It's inevitable, because individual cores just can't get much faster. At some point, that's going to mean that we have to rethink our architecture entirely, as millions of cores can't all access a shared memory space efficiently. But millions of cores are still a long way off, and in the meantime we'll see machines with dozens of cores, struggling with shared memory. Alex's tip: The best way for an application to make use of that increasing parallel power is to use a concurrency model like actors, that deals with synchronisation issues for you. Then, the maintainer of the actors framework can find the most efficient way to coordinate access to shared memory to allow your actors to pass messages to each other efficiently. At the moment, NAct uses the .NET thread pool and a few locks to marshal messages. It works well on dual and quad core machines, but it won't scale to more cores. Every time we use a lock, our core performs an atomic memory operation (eg. CAS) on a cell of memory representing the lock, so it's sure that no other core can possibly have that lock. This is very fast when the lock isn't contended, but we need to notify all the other cores, in case they held the cell of memory in a cache. As the number of cores increases, the total cost of a lock increases linearly. A lot of work has been done on "lock-free" data structures, which avoid locks by using atomic memory operations directly. These give fairly dramatic performance improvements, particularly on systems with a few (2 to 4) cores. The .NET 4 concurrent collections in System.Collections.Concurrent are mostly lock-free. However, lock-free data structures still don't scale indefinitely, because any use of an atomic memory operation still involves every core in the system. A sync-free data structure Some concurrent data structures are possible to write in a completely synchronization-free way, without using any atomic memory operations. One useful example is a single producer, single consumer (SPSC) queue. It's easy to write a sync-free fixed size SPSC queue using a circular buffer*. Slightly trickier is a queue that grows as needed. You can use a linked list to represent the queue, but if you leave the nodes to be garbage collected once you're done with them, the GC will need to involve all the cores in collecting the finished nodes. Instead, I've implemented a proof of concept inspired by this intel article which reuses the nodes by putting them in a second queue to send back to the producer. * In all these cases, you need to use memory barriers correctly, but these are local to a core, so don't have the same scalability problems as atomic memory operations. Performance tests I tried benchmarking my SPSC queue against the .NET ConcurrentQueue, and against a standard Queue protected by locks. In some ways, this isn't a fair comparison, because both of these support multiple producers and multiple consumers, but I'll come to that later. I started on my dual-core laptop, running a simple test that had one thread producing 64 bit integers, and another consuming them, to measure the pure overhead of the queue. So, nothing very interesting here. Both concurrent collections perform better than the lock-based one as expected, but there's not a lot to choose between the ConcurrentQueue and my SPSC queue. I was a little disappointed, but then, the .NET Framework team spent a lot longer optimising it than I did. So I dug out a more powerful machine that Red Gate's DBA tools team had been using for testing. It is a 6 core Intel i7 machine with hyperthreading, adding up to 12 logical cores. Now the results get more interesting. As I increased the number of producer-consumer pairs to 6 (to saturate all 12 logical cores), the locking approach was slow, and got even slower, as you'd expect. What I didn't expect to be so clear was the drop-off in performance of the lock-free ConcurrentQueue. I could see the machine only using about 20% of available CPU cycles when it should have been saturated. My interpretation is that as all the cores used atomic memory operations to safely access the queue, they ended up spending most of the time notifying each other about cache lines that need invalidating. The sync-free approach scaled perfectly, despite still working via shared memory, which after all, should still be a bottleneck. I can't quite believe that the results are so clear, so if you can think of any other effects that might cause them, please comment! Obviously, this benchmark isn't realistic because we're only measuring the overhead of the queue. Any real workload, even on a machine with 12 cores, would dwarf the overhead, and there'd be no point worrying about this effect. But would that be true on a machine with 100 cores? Still to be solved. The trouble is, you can't build many concurrent algorithms using only an SPSC queue to communicate. In particular, I can't see a way to build something as general purpose as actors on top of just SPSC queues. Fundamentally, an actor needs to be able to receive messages from multiple other actors, which seems to need an MPSC queue. I've been thinking about ways to build a sync-free MPSC queue out of multiple SPSC queues and some kind of sign-up mechanism. Hopefully I'll have something to tell you about soon, but leave a comment if you have any ideas.

  Read the article

• ASSIMP in my program is much slower to import than ASSIMP view program

  - by Marco

  The problem is really simple: if I try to load with the function aiImportFileExWithProperties a big model in my software (around 200.000 vertices), it takes more than one minute. If I try to load the very same model with ASSIMP view, it takes 2 seconds. For this comparison, both my software and Assimp view are using the dll version of the library at 64 bit, compiled by myself (Assimp64.dll). This is the relevant piece of code in my software // default pp steps unsigned int ppsteps = aiProcess_CalcTangentSpace | // calculate tangents and bitangents if possible aiProcess_JoinIdenticalVertices | // join identical vertices/ optimize indexing aiProcess_ValidateDataStructure | // perform a full validation of the loader's output aiProcess_ImproveCacheLocality | // improve the cache locality of the output vertices aiProcess_RemoveRedundantMaterials | // remove redundant materials aiProcess_FindDegenerates | // remove degenerated polygons from the import aiProcess_FindInvalidData | // detect invalid model data, such as invalid normal vectors aiProcess_GenUVCoords | // convert spherical, cylindrical, box and planar mapping to proper UVs aiProcess_TransformUVCoords | // preprocess UV transformations (scaling, translation ...) aiProcess_FindInstances | // search for instanced meshes and remove them by references to one master aiProcess_LimitBoneWeights | // limit bone weights to 4 per vertex aiProcess_OptimizeMeshes | // join small meshes, if possible; aiProcess_SplitByBoneCount | // split meshes with too many bones. Necessary for our (limited) hardware skinning shader 0; cout << "Loading " << pFile << "... "; aiPropertyStore* props = aiCreatePropertyStore(); aiSetImportPropertyInteger(props,AI_CONFIG_IMPORT_TER_MAKE_UVS,1); aiSetImportPropertyFloat(props,AI_CONFIG_PP_GSN_MAX_SMOOTHING_ANGLE,80.f); aiSetImportPropertyInteger(props,AI_CONFIG_PP_SBP_REMOVE, aiPrimitiveType_LINE | aiPrimitiveType_POINT); aiSetImportPropertyInteger(props,AI_CONFIG_GLOB_MEASURE_TIME,1); //aiSetImportPropertyInteger(props,AI_CONFIG_PP_PTV_KEEP_HIERARCHY,1); // Call ASSIMPs C-API to load the file scene = (aiScene*)aiImportFileExWithProperties(pFile.c_str(), ppsteps | /* default pp steps */ aiProcess_GenSmoothNormals | // generate smooth normal vectors if not existing aiProcess_SplitLargeMeshes | // split large, unrenderable meshes into submeshes aiProcess_Triangulate | // triangulate polygons with more than 3 edges //aiProcess_ConvertToLeftHanded | // convert everything to D3D left handed space aiProcess_SortByPType | // make 'clean' meshes which consist of a single typ of primitives 0, NULL, props); aiReleasePropertyStore(props); if(!scene){ cout << aiGetErrorString() << endl; return 0; } this is the relevant piece of code in assimp view code // default pp steps unsigned int ppsteps = aiProcess_CalcTangentSpace | // calculate tangents and bitangents if possible aiProcess_JoinIdenticalVertices | // join identical vertices/ optimize indexing aiProcess_ValidateDataStructure | // perform a full validation of the loader's output aiProcess_ImproveCacheLocality | // improve the cache locality of the output vertices aiProcess_RemoveRedundantMaterials | // remove redundant materials aiProcess_FindDegenerates | // remove degenerated polygons from the import aiProcess_FindInvalidData | // detect invalid model data, such as invalid normal vectors aiProcess_GenUVCoords | // convert spherical, cylindrical, box and planar mapping to proper UVs aiProcess_TransformUVCoords | // preprocess UV transformations (scaling, translation ...) aiProcess_FindInstances | // search for instanced meshes and remove them by references to one master aiProcess_LimitBoneWeights | // limit bone weights to 4 per vertex aiProcess_OptimizeMeshes | // join small meshes, if possible; aiProcess_SplitByBoneCount | // split meshes with too many bones. Necessary for our (limited) hardware skinning shader 0; aiPropertyStore* props = aiCreatePropertyStore(); aiSetImportPropertyInteger(props,AI_CONFIG_IMPORT_TER_MAKE_UVS,1); aiSetImportPropertyFloat(props,AI_CONFIG_PP_GSN_MAX_SMOOTHING_ANGLE,g_smoothAngle); aiSetImportPropertyInteger(props,AI_CONFIG_PP_SBP_REMOVE,nopointslines ? aiPrimitiveType_LINE | aiPrimitiveType_POINT : 0 ); aiSetImportPropertyInteger(props,AI_CONFIG_GLOB_MEASURE_TIME,1); //aiSetImportPropertyInteger(props,AI_CONFIG_PP_PTV_KEEP_HIERARCHY,1); // Call ASSIMPs C-API to load the file g_pcAsset->pcScene = (aiScene*)aiImportFileExWithProperties(g_szFileName, ppsteps | /* configurable pp steps */ aiProcess_GenSmoothNormals | // generate smooth normal vectors if not existing aiProcess_SplitLargeMeshes | // split large, unrenderable meshes into submeshes aiProcess_Triangulate | // triangulate polygons with more than 3 edges aiProcess_ConvertToLeftHanded | // convert everything to D3D left handed space aiProcess_SortByPType | // make 'clean' meshes which consist of a single typ of primitives 0, NULL, props); aiReleasePropertyStore(props); As you can see the code is nearly identical because I copied from assimp view. What could be the reason for such a difference in performance? The two software are using the same dll Assimp64.dll (compiled in my computer with vc++ 2010 express) and the same function aiImportFileExWithProperties to load the model, so I assume that the actual code employed is the same. How is it possible that the function aiImportFileExWithProperties is 100 times slower when called by my sotware than when called by assimp view? What am I missing? I am not good with dll, dynamic and static libraries so I might be missing something obvious. ------------------------------ UPDATE I found out the reason why the code is going slower. Basically I was running my software with "Start debugging" in VC++ 2010 Express. If I run the code outside VC++ 2010 I get same performance of assimp view. However now I have a new question. Why does the dll perform slower in VC++ debugging? I compiled it in release mode without debugging information. Is there any way to have the dll go fast in debugmode i.e. not debugging the dll? Because I am interested in debugging only my own code, not the dll that I assume is already working fine. I do not want to wait 2 minutes every time I want to load my software to debug. Does this request make sense?

  Read the article

• T-SQL Tuesday #21 - Crap!

  - by Most Valuable Yak (Rob Volk)

  Adam Machanic's (blog | twitter) ever popular T-SQL Tuesday series is being held on Wednesday this time, and the topic is… SHIT CRAP. No, not fecal material.  But crap code.  Crap SQL.  Crap ideas that you thought were good at the time, or were forced to do due (doo-doo?) to lack of time. The challenge for me is to look back on my SQL Server career and find something that WASN'T crap.  Well, there's a lot that wasn't, but for some reason I don't remember those that well.  So the additional challenge is to pick one particular turd that I really wish I hadn't squeezed out.  Let's see if this outline fits the bill: An ETL process on text files; That had to interface between SQL Server and an AS/400 system; That didn't use SSIS (should have) or BizTalk (ummm, no) but command-line scripting, using Unix utilities(!) via: xp_cmdshell; That had to email reports and financial data, some of it sensitive Yep, the stench smell is coming back to me now, as if it was yesterday… As to why SSIS and BizTalk were not options, basically I didn't know either of them well enough to get the job done (and I still don't).  I also had a strict deadline of 3 days, in addition to all the other responsibilities I had, so no time to learn them.  And seeing how screwed up the rest of the process was: Payment files from multiple vendors in multiple formats; Sent via FTP, PGP encrypted email, or some other wizardry; Manually opened/downloaded and saved to a particular set of folders (couldn't change this); Once processed, had to be placed BACK in the same folders with the original archived; x2 divisions that had to run separately; Plus an additional vendor file in another format on a completely different schedule; So that they could be MANUALLY uploaded into the AS/400 system (couldn't change this either, even if it was technically possible) I didn't feel so bad about the solution I came up with, which was naturally: Copy the payment files to the local SQL Server drives, using xp_cmdshell Run batch files (via xp_cmdshell) to parse the different formats using sed, a Unix utility (this was before Powershell) Use other Unix utilities (join, split, grep, wc) to process parsed files and generate metadata (size, date, checksum, line count) Run sqlcmd to execute a stored procedure that passed the parsed file names so it would bulk load the data to do a comparison bcp the compared data out to ANOTHER text file so that I could grep that data out of the original file Run another stored procedure to import the matched data into SQL Server so it could process the payments, including file metadata Process payment batches and log which division and vendor they belong to Email the payment details to the finance group (since it was too hard for them to run a web report with the same data…which they ran anyway to compare the emailed file against…which always matched, surprisingly) Email another report showing unmatched payments so they could manually void them…about 3 months afterward All in "Excel" format, using xp_sendmail (SQL 2000 system) Copy the unmatched data back to the original folder locations, making sure to match the file format exactly (if you've ever worked with ACH files, you'll understand why this sucked) If you're one of the 10 people who have read my blog before, you know that I love the DOS "for" command.  Like passionately.  Like fairy-tale love.  So my batch files were riddled with for loops, nested within other for loops, that called other batch files containing for loops.  I think there was one section that had 4 or 5 nested for commands.  It was wrong, disturbed, and completely un-maintainable by anyone, even myself.  Months, even a year, after I left the company I got calls from someone who had to make a minor change to it, and they called me to talk them out of spraying the office with an AK-47 after looking at this code.  (for you Star Trek TOS fans) The funniest part of this, well, one of the funniest, is that I made the deadline…sort of, I was only a day late…and the DAMN THING WORKED practically unchanged for 3 years.  Most of the problems came from the manual parts of the overall process, like forgetting to decrypt the files, or missing/late files, or saved to the wrong folders.  I'm definitely not trying to toot my own horn here, because this was truly one of the dumbest, crappiest solutions I ever came up with.  Fortunately as far as I know it's no longer in use and someone has written a proper replacement.  Today I would knuckle down and do it in SSIS or Powershell, even if it took me weeks to get it right. The real lesson from this crap code is to make things MAINTAINABLE and UNDERSTANDABLE.  sed scripting regular expressions doesn't fit that criteria in any way.  If you ever find yourself under pressure to do something fast at all costs, DON'T DO IT.  Stop and consider long-term maintainability, not just for yourself but for others on your team.  If you can't explain the basic approach in under 5 minutes, it ultimately won't succeed.  And while you may love to leave all that crap behind, it may follow you anyway, and you'll step in it again.   P.S. - if you're wondering about all the manual stuff that couldn't be changed, it was because the entire process had gone through Six Sigma, and was deemed the best possible way.  Phew!  Talk about stink!

  Read the article

• Oracle Social Network Developer Challenge: Bezzotech

  - by Kellsey Ruppel

  Originally posted by Jake Kuramoto on The Apps Lab blog. I’ve covered all the entries we had for the Oracle Social Network Developer Challenge, the winners, Dimitri and Martin, HarQen, TEAM Informatics and John Sim from Fishbowl Solutions, and today, I’m giving you bonus coverage. Friend of the ‘Lab, Bex Huff (@bex) from Bezzotech (@bezzotech), had an interesting OpenWorld. He rebounded from an allergic reaction to finish his entry, Honey Badger, only to have his other OpenWorld commitments make him unable to present his work. Still, he did a bunch of work, and I want to make sure everyone knows about the Honey Badger. If you’re wondering about the name, it’s a meme; “honey badger don’t care.” Bex tackled a common problem with social tools by adding game mechanics to create an incentive for people to keep their profiles updated. He used a Hot-or-Not style comparison app that poses expertise questions and awards a badge to the winner. Questions are based on whatever attributes the business wants to emphasize. The goal is to find the mavens in an organization, give them praise and recognition, ideally creating incentive for everyone to raise their games. In his own words: There is a real information quality problem in social networks. In last year’s keynote, Larry Elison demonstrated how to use the social network to track down resources that have the skill sets needed for specific projects. But how well would that work in real life? People usually update that information with the basic profile information, but they rarely update their profiles with latest news items, projects, customers, or skills. It’s a pain. Or, put another way, when was the last time you updated your LinkedIn profile? Enter the Honey Badger! This is a example of a comparator app that gamifies the way people keep their profiles updated, which ensures higher quality data in the social network. An administrator comes up with a series of important questions: Who is a better communicator? Who is a better Java programmer? Who is a better team player? And people would have a space in their profile to give a justification as to why they have these skills. The second part of the app is the comparator. It randomly shows two people, their names, and their justification for why they have these skills. You will click on one of them to “vote” for them, then on the next page you will see the results from the previous match, and get 2 new people to vote on. Anybody with a winning score wins a “Honey Badge” to be displayed on their profile page, which proudly states that their peers agree that this person has those skills. Once a badge is won, it will be jealously guarded. The longer your go without updating your profile, the more likely it is that you will lose your badge. This “loss aversion” is well known in psychology, and is a strong incentive for people to keep their profiles up to date. If a user sees their rank drop from 90% to 60%, they will find the time to update their justification! Unfortunately, during the hackathon we were not allowed to modify the schema to allow for additional fields such as “justification.” So this hack is limited to just the one basic question: who is the bigger Honey Badger? Here are some shots of the Honey Badger application: #gallery-1 { margin: auto; } #gallery-1 .gallery-item { float: left; margin-top: 10px; text-align: center; width: 33%; } #gallery-1 img { border: 2px solid #cfcfcf; } #gallery-1 .gallery-caption { margin-left: 0; } Thanks to Bex and everyone for participating in our challenge. Despite very little time to promote this event, we had a great turnout and creative and useful entries. The amount of work required to put together these final entries was significant, especially during a conference, and the judges and all of us involved were impressed at how much work everyone was able to do. Congrats to everyone, pat yourselves on the back. Stay tuned if you’re interested in challenges like these. We’ll likely be running similar events in the not-so-distant future.

  Read the article

• Using SQL Source Control with Fortress or Vault &ndash; Part 1

  - by AjarnMark

  I am fanatical when it comes to managing the source code for my company.  Everything that we build (in source form) gets put into our source control management system.  And I’m not just talking about the UI and middle-tier code written in C# and ASP.NET, but also the back-end database stuff, which at times has been a pain.  We even script out our Scheduled Jobs and keep a copy of those under source control. The UI and middle-tier stuff has long been easy to manage as we mostly use Visual Studio which has integration with source control systems built in.  But the SQL code has been a little harder to deal with.  I have been doing this for many years, well before Microsoft came up with Data Dude, so I had already established a methodology that, while not as smooth as VS, nonetheless let me keep things well controlled, and allowed doing my database development in my tool of choice, Query Analyzer in days gone by, and now SQL Server Management Studio.  It just makes sense to me that if I’m going to do database development, let’s use the database tool set.  (Although, I have to admit I was pretty impressed with the demo of Juneau that Don Box did at the PASS Summit this year.)  So as I was saying, I had developed a methodology that worked well for us (and I’ll probably outline in a future post) but it could use some improvement. When Solutions and Projects were first introduced in SQL Management Studio, I thought we were finally going to get our same experience that we have in Visual Studio.  Well, let’s say I was underwhelmed by Version 1 in SQL 2005, and apparently so were enough other people that by the time SQL 2008 came out, Microsoft decided that Solutions and Projects would be deprecated and completely removed from a future version.  So much for that idea. Then I came across SQL Source Control from Red-Gate.  I have used several tools from Red-Gate in the past, including my favorites SQL Compare, SQL Prompt, and SQL Refactor.  SQL Prompt is worth its weight in gold, and the others are great, too.  Earlier this year, we upgraded from our earlier product bundles to the new Developer Bundle, and in the process added SQL Source Control to our collection.  I thought this might really be the golden ticket I was looking for.  But my hopes were quickly dashed when I discovered that it only integrated with Microsoft Team Foundation Server and Subversion as the source code repositories.  We have been using SourceGear’s Vault and Fortress products for years, and I wholeheartedly endorse them.  So I was out of luck for the time being, although there were a number of people voting for Vault/Fortress support on their feedback forum (as did I) so I had hope that maybe next year I could look at it again. But just a couple of weeks ago, I was pleasantly surprised to receive notice in my email that Red-Gate had an Early Access version of SQL Source Control that worked with Vault and Fortress, so I quickly downloaded it and have been putting it through its paces.  So far, I really like what I see, and I have been quite impressed with Red-Gate’s responsiveness when I have contacted them with any issues or concerns that I have had.  I have had several communications with Gyorgy Pocsi at Red-Gate and he has been immensely helpful and responsive. I must say that development with SQL Source Control is very different from what I have been used to.  This post is getting long enough, so I’ll save some of the details for a separate write-up, but the short story is that in my regular mode, it’s all about the script files.  Script files are King and you dare not make a change to the database other than by way of a script file, or you are in deep trouble.  With SQL Source Control, you make your changes to your development database however you like.  I still prefer writing most of my changes in T-SQL, but you can also use any of the GUI functionality of SSMS to make your changes, and SQL Source Control “manages” the script for you.  Basically, when you first link your database to source control, the tool generates scripts for every primary object (tables and their indexes are together in one script, not broken out into separate scripts like DB Projects do) and those scripts are checked into your source control.  So, if you needed to, you could still do a GET from your source control repository and build the database from scratch.  But for the day-to-day work, SQL Source Control uses the same technique as SQL Compare to determine what changes have been made to your development database and how to represent those in your repository scripts.  I think that once I retrain myself to just work in the database and quit worrying about having to find and open the right script file, that this will actually make us more efficient. And for deployment purposes, SQL Source Control integrates with the full SQL Compare utility to produce a synchronization script (or do a live sync).  This is similar in concept to Microsoft’s DACPAC, if you’re familiar with that. If you are not currently keeping your database development efforts under source control, definitely examine this tool.  If you already have a methodology that is working for you, then I still think this is worth a review and comparison to your current approach.  You may find it more efficient.  But remember that the version which integrates with Vault/Fortress is still in pre-release mode, so treat it with a little caution.  I have found it to be fairly stable, but there was one bug that I found which had inconvenient side-effects and could have really been frustrating if I had been running this on my normal active development machine.  However, I can verify that that bug has been fixed in a more recent build version (did I mention Red-Gate’s responsiveness?).

  Read the article

• MySQL Connect 9 Days Away – Optimizer Sessions

  - by Bertrand Matthelié

  72 1024x768 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:0cm 5.4pt 0cm 5.4pt; mso-para-margin:0cm; 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;} Following my previous blog post focusing on InnoDB talks at MySQL Connect, let us review today the sessions focusing on the MySQL Optimizer: Saturday, 11.30 am, Room Golden Gate 6: MySQL Optimizer Overview—Olav Sanstå, Oracle The goal of MySQL optimizer is to take a SQL query as input and produce an optimal execution plan for the query. This session presents an overview of the main phases of the MySQL optimizer and the primary optimizations done to the query. These optimizations are based on a combination of logical transformations and cost-based decisions. Examples of optimization strategies the presentation covers are the main query transformations, the join optimizer, the data access selection strategies, and the range optimizer. For the cost-based optimizations, an overview of the cost model and the data used for doing the cost estimations is included. Saturday, 1.00 pm, Room Golden Gate 6: Overview of New Optimizer Features in MySQL 5.6—Manyi Lu, Oracle Many optimizer features have been added into MySQL 5.6. This session provides an introduction to these great features. Multirange read, index condition pushdown, and batched key access will yield huge performance improvements on large data volumes. Structured explain, explain for update/delete/insert, and optimizer tracing will help users analyze and speed up queries. And last but not least, the session covers subquery optimizations in Release 5.6. Saturday, 7.00 pm, Room Golden Gate 4: BoF: Query Optimizations: What Is New and What Is Coming? This BoF presents common techniques for query optimization, covers what is new in MySQL 5.6, and provides a discussion forum in which attendees can tell the MySQL optimizer team which optimizations they would like to see in the future. Sunday, 1.15 pm, Room Golden Gate 8: Query Performance Comparison of MySQL 5.5 and MySQL 5.6—Øystein Grøvlen, Oracle MySQL Release 5.6 contains several improvements in the query optimizer that create improved performance for complex queries. This presentation looks at how MySQL 5.6 improves the performance of many of the queries in the DBT-3 benchmark. Based on the observed improvements, the presentation discusses what makes the specific queries perform better in Release 5.6. It describes the relevant new optimization techniques and gives examples of the types of queries that will benefit from these techniques. Sunday, 4.15 pm, Room Golden Gate 4: Powerful EXPLAIN in MySQL 5.6—Evgeny Potemkin, Oracle The EXPLAIN command of MySQL has long been a very useful tool for understanding how MySQL will execute a query. Release 5.6 of the MySQL database offers several new additions that give more-detailed information about the query plan and make it easier to understand at the same time. This presentation gives an overview of new EXPLAIN features: structured EXPLAIN in JSON format, EXPLAIN for INSERT/UPDATE/DELETE, and optimizer tracing. Examples in the session give insights into how you can take advantage of the new features. They show how these features supplement and relate to each other and to classical EXPLAIN and how and why the MySQL server chooses a particular query plan. You can check out the full program here as well as in the September edition of the MySQL newsletter. Not registered yet? You can still save US$ 300 over the on-site fee – Register Now!

  Read the article

• IBM "per core" comparisons for SPECjEnterprise2010

  - by jhenning

  I recently stumbled upon a blog entry from Roman Kharkovski (an IBM employee) comparing some SPECjEnterprise2010 results for IBM vs. Oracle. Mr. Kharkovski's blog claims that SPARC delivers half the transactions per core vs. POWER7. Prior to any argument, I should say that my predisposition is to like Mr. Kharkovski, because he says that his blog is intended to be factual; that the intent is to try to avoid marketing hype and FUD tactic; and mostly because he features a picture of himself wearing a bike helmet (me too). Therefore, in a spirit of technical argument, rather than FUD fight, there are a few areas in his comparison that should be discussed. Scaling is not free For any benchmark, if a small system scores 13k using quantity R1 of some resource, and a big system scores 57k using quantity R2 of that resource, then, sure, it's tempting to divide: is  13k/R1 > 57k/R2 ? It is tempting, but not necessarily educational. The problem is that scaling is not free. Building big systems is harder than building small systems. Scoring  13k/R1  on a little system provides no guarantee whatsoever that one can sustain that ratio when attempting to handle more than 4 times as many users. Choosing the denominator radically changes the picture When ratios are used, one can vastly manipulate appearances by the choice of denominator. In this case, lots of choices are available for the resource to be compared (R1 and R2 above). IBM chooses to put cores in the denominator. Mr. Kharkovski provides some reasons for that choice in his blog entry. And yet, it should be noted that the very concept of a core is: arbitrary: not necessarily comparable across vendors; fluid: modern chips shift chip resources in response to load; and invisible: unless you have a microscope, you can't see it. By contrast, one can actually see processor chips with the naked eye, and they are a bit easier to count. If we put chips in the denominator instead of cores, we get: 13161.07 EjOPS / 4 chips = 3290 EjOPS per chip for IBM vs 57422.17 EjOPS / 16 chips = 3588 EjOPS per chip for Oracle The choice of denominator makes all the difference in the appearance. Speaking for myself, dividing by chips just seems to make more sense, because: I can see chips and count them; and I can accurately compare the number of chips in my system to the count in some other vendor's system; and Tthe probability of being able to continue to accurately count them over the next 10 years of microprocessor development seems higher than the probability of being able to accurately and comparably count "cores". SPEC Fair use requirements Speaking as an individual, not speaking for SPEC and not speaking for my employer, I wonder whether Mr. Kharkovski's blog article, taken as a whole, meets the requirements of the SPEC Fair Use rule www.spec.org/fairuse.html section I.D.2. For example, Mr. Kharkovski's footnote (1) begins Results from http://www.spec.org as of 04/04/2013 Oracle SUN SPARC T5-8 449 EjOPS/core SPECjEnterprise2010 (Oracle's WLS best SPECjEnterprise2010 EjOPS/core result on SPARC). IBM Power730 823 EjOPS/core (World Record SPECjEnterprise2010 EJOPS/core result) The questionable tactic, from a Fair Use point of view, is that there is no such metric at the designated location. At www.spec.org, You can find the SPEC metric 57422.17 SPECjEnterprise2010 EjOPS for Oracle and You can also find the SPEC metric 13161.07 SPECjEnterprise2010 EjOPS for IBM. Despite the implication of the footnote, you will not find any mention of 449 nor anything that says 823. SPEC says that you can, under its fair use rule, derive your own values; but it emphasizes: "The context must not give the appearance that SPEC has created or endorsed the derived value." Substantiation and transparency Although SPEC disclaims responsibility for non-SPEC information (section I.E), it says that non-SPEC data and methods should be accurate, should be explained, should be substantiated. Unfortunately, it is difficult or impossible for the reader to independently verify the pricing: Were like units compared to like (e.g. list price to list price)? Were all components (hw, sw, support) included? Were all fees included? Note that when tpc.org shows IBM pricing, there are often items such as "PROCESSOR ACTIVATION" and "MEMORY ACTIVATION". Without the transparency of a detailed breakdown, the pricing claims are questionable. T5 claim for "Fastest Processor" Mr. Kharkovski several times questions Oracle's claim for fastest processor, writing You see, when you publish industry benchmarks, people may actually compare your results to other vendor's results. Well, as we performance people always say, "it depends". If you believe in performance-per-core as the primary way of looking at the world, then yes, the POWER7+ is impressive, spending its chip resources to support up to 32 threads (8 cores x 4 threads). Or, it just might be useful to consider performance-per-chip. Each SPARC T5 chip allows 128 hardware threads to be simultaneously executing (16 cores x 8 threads). The Industry Standard Benchmark that focuses specifically on processor chip performance is SPEC CPU2006. For this very well known and popular benchmark, SPARC T5: provides better performance than both POWER7 and POWER7+, for 1 chip vs. 1 chip, for 8 chip vs. 8 chip, for integer (SPECint_rate2006) and floating point (SPECfp_rate2006), for Peak tuning and for Base tuning. For example, at the 8-chip level, integer throughput (SPECint_rate2006) is: 3750 for SPARC 2170 for POWER7+. You can find the details at the March 2013 BestPerf CPU2006 page SPEC is a trademark of the Standard Performance Evaluation Corporation, www.spec.org. The two specific results quoted for SPECjEnterprise2010 are posted at the URLs linked from the discussion. Results for SPEC CPU2006 were verified at spec.org 1 July 2013, and can be rechecked here.

  Read the article

• First time shadow mapping problems

  - by user1294203

  I have implemented basic shadow mapping for the first time in OpenGL using shaders and I'm facing some problems. Below you can see an example of my rendered scene: The process of the shadow mapping I'm following is that I render the scene to the framebuffer using a View Matrix from the light point of view and the projection and model matrices used for normal rendering. In the second pass, I send the above MVP matrix from the light point of view to the vertex shader which transforms the position to light space. The fragment shader does the perspective divide and changes the position to texture coordinates. Here is my vertex shader, #version 150 core uniform mat4 ModelViewMatrix; uniform mat3 NormalMatrix; uniform mat4 MVPMatrix; uniform mat4 lightMVP; uniform float scale; in vec3 in_Position; in vec3 in_Normal; in vec2 in_TexCoord; smooth out vec3 pass_Normal; smooth out vec3 pass_Position; smooth out vec2 TexCoord; smooth out vec4 lightspace_Position; void main(void){ pass_Normal = NormalMatrix * in_Normal; pass_Position = (ModelViewMatrix * vec4(scale * in_Position, 1.0)).xyz; lightspace_Position = lightMVP * vec4(scale * in_Position, 1.0); TexCoord = in_TexCoord; gl_Position = MVPMatrix * vec4(scale * in_Position, 1.0); } And my fragment shader, #version 150 core struct Light{ vec3 direction; }; uniform Light light; uniform sampler2D inSampler; uniform sampler2D inShadowMap; smooth in vec3 pass_Normal; smooth in vec3 pass_Position; smooth in vec2 TexCoord; smooth in vec4 lightspace_Position; out vec4 out_Color; float CalcShadowFactor(vec4 lightspace_Position){ vec3 ProjectionCoords = lightspace_Position.xyz / lightspace_Position.w; vec2 UVCoords; UVCoords.x = 0.5 * ProjectionCoords.x + 0.5; UVCoords.y = 0.5 * ProjectionCoords.y + 0.5; float Depth = texture(inShadowMap, UVCoords).x; if(Depth < (ProjectionCoords.z + 0.001)) return 0.5; else return 1.0; } void main(void){ vec3 Normal = normalize(pass_Normal); vec3 light_Direction = -normalize(light.direction); vec3 camera_Direction = normalize(-pass_Position); vec3 half_vector = normalize(camera_Direction + light_Direction); float diffuse = max(0.2, dot(Normal, light_Direction)); vec3 temp_Color = diffuse * vec3(1.0); float specular = max( 0.0, dot( Normal, half_vector) ); float shadowFactor = CalcShadowFactor(lightspace_Position); if(diffuse != 0 && shadowFactor > 0.5){ float fspecular = pow(specular, 128.0); temp_Color += fspecular; } out_Color = vec4(shadowFactor * texture(inSampler, TexCoord).xyz * temp_Color, 1.0); } One of the problems is self shadowing as you can see in the picture, the crate has its own shadow cast on itself. What I have tried is enabling polygon offset (i.e. glEnable(POLYGON_OFFSET_FILL), glPolygonOffset(GLfloat, GLfloat) ) but it didn't change much. As you see in the fragment shader, I have put a static offset value of 0.001 but I have to change the value depending on the distance of the light to get more desirable effects , which not very handy. I also tried using front face culling when I render to the framebuffer, that didn't change much too. The other problem is that pixels outside the Light's view frustum get shaded. The only object that is supposed to be able to cast shadows is the crate. I guess I should pick more appropriate projection and view matrices, but I'm not sure how to do that. What are some common practices, should I pick an orthographic projection? From googling around a bit, I understand that these issues are not that trivial. Does anyone have any easy to implement solutions to these problems. Could you give me some additional tips? Please ask me if you need more information on my code. Here is a comparison with and without shadow mapping of a close-up of the crate. The self-shadowing is more visible.

  Read the article

• Windows Phone 7 Review &ndash; Part 1: LG Quantum

  - by Nikita Polyakov

  As many of my fellow geeks, I ran out and got a retail windows Phone 7 on the first day. Just had to have it :) I’ve had the developer prototypes in my hands for previous 3 months on and off, so I finally wanted to have one I call my own. I’ve rushed the Launch   I’ve checked out both AT&T and T-Mobile offerings on day 1 and decided on a Samsung Focus. Great screen, super light and thin. If you don’t believe me that this phone can compete with the best of the non-Phone 7 offerings - get it in your hand to compare for yourself. I have to say that even though the on-screen keyboard on Windows Phone 7 is one of the best, the amount of text I write on my phone and my expectation of how long that takes for a short reply are very high. Also the phone being so slick and sexy did not feel solid or confident in my hand or pocket. As the dust settled   Arrives the LG Quantum – now on AT&T and worldwide. First impression of the softer plastic, the back battery cover is solid metal - the entire phone feels solid and indestructible! Phone fits just right in my hand, it’s almost too good. It does not feel like it will crack in your jeans. I feel safe holding it and don’t feel like if I or someone were to bump into me walking it’d fly out of my hand. I’ve dropped and had thrown the Focus a few times on accident as it’s weight is negligible. I won’t even dream of lying the first day adjusting to a 3.5’ LCD screen from the Samsung’s blistering bright and poppy AMOLED 4’ was hard. But the colors and sharpness are still very good. I find it almost easier on the eyes actually for day to day use.  I had a chance to lay the phone down in the line with the prototypes and final versions of other phones that had LCD screens – LG makes HTC looks like a budget LCD compared to a high end LCD in the home theatre department. I am consistently complemented by friends that have the HD7 or Surround on how much better my screen looks. The screen just looks like the most color correct phone out of the line up. Even next to Samsung it makes it look oversaturated, but can’t match the true blacks compensating with true white.   Day to Day Usability   What I also noticed that is a huge difference is how much I am not accidently hitting the soft keys at the bottom. I real pain on Focus since holding it in am average size hand already would accidently touch the controls at the bottom. QWERTY keyboard on this phone is great. It’s like the mission for LG is “make it solid!”. Keyboard has a very durable feel.   LG’s has a secret wild card though is the DLNA support. If you seen an ad for it, you should. Imagine this – playing a song from your phone straight to your network connected A/V receiver. Done. Pictures to TV. Done. Video. Done. DLNA works with components that advertise to as well as Windows 7, XBOX 360 and other consoles.  I will write an extensive review of that experience in near future. LG Exclusive apps – from panorama photo taker to voice to text translator and even look-n-type app that works like a backup inverse camera, there is quite a bit there that won’t be found on the other phones. I’ll review those in more detail in another segment. Conclusion So for a quick comparison: If you want a phone that is super thin, light and is core reference of a Windows Phone 7 – Samsung Focus it is. If you want a great phone with solid secure feel, real keyboard, media features - the hands down winner is LG Quantum.   You can pick up the LG Quantum at AT&T in US and worldwide as LG Optimus 7Q.   Final thought: I have not had SmartPhone that I felt was a reliable trusty primary communication device since Samsung BlackJack II, this time the LG got the crown.   [ Disclosure: Phone was provided to me free of charge. That has been the case for all of my phones for years, nothing new - I get them all. ]

  Read the article

• Community Branching

  - by Dane Morgridge

  As some may have noticed, I have taken a liking to Ruby (and Rails in particular) quite a bit recently. This last weekend I spoke at the NYC Code Camp on a comparison of ASP.NET and Rails as well as an intro to Entity Framework talk.  I am speaking at RubyNation in April and have submitted to other ruby conferences around the area and I am also doing a Rails and MongoDB talk at the Philly Code Camp in April. Before you start to think this is my "I'm leaving .NET post", which it isn't so I need to clarify. I am not, nor do I intend to any time in the near future plan on abandoning .NET.  I am simply branching out into another community based on a development technology that I very much enjoy.  If you look at my twitter bio, you will see that I am into Entity Framework, Ruby on Rails, C++ and ASP.NET MVC, and not necessarily in that order.  I know you're probably thinking to your self that I am crazy, which is probably true on several levels (especially the C++ part). I was actually crazy enough at the NYC Code Camp to show up wearing a Linux t-shirt, presenting with my MacBook Pro on Entity Framework, ASP.NET MVC and Rails. (I did get pelted in the head with candy by Rachel Appel for it though) At all of the code camps I am submitting to this year, i will be submitting sessions on likely all four topics, and some sessions will be a combination of 2 or more.  For example, my "ASP.NET MVC: A Gateway To Rails?" talk touches ASP.NET MVC, Entity Framework Code First and Rails. Simply put (and I talk about this in my MVC & Rails talk) is that learning and using Rails has made me a better ASP.NET MVC developer. Just one example of this is helper methods.  When I started working with ASP.NET MVC, I didn't really want to use helpers and preferred to just use standard html tags, especially where links were concerned.  It was just me being stubborn and not really seeing all of the benefit of the helpers.  To my defense, coming from WebForms, I wanted to be as bare metal as possible and it seemed at first like a lot of the helpers were an unnecessary abstraction. I took my first look at Rails back in v1 and didn't spend very much time with it so I dismissed it and went on my merry ASP.NET WebForms way.  Then I picked up ASP.NET MVC and grasped the MVC pattern itself much better. After this, I took another look at Rails and everything made sense.  I decided then to learn Rails. (I think it is important for developers to learn new languages and platforms regularly so it was a natural progression for me) I wanted to learn it the right way, so when I dug into code, everyone used helpers everywhere for pretty much everything possible. I took some time to dig in and found out how helpful they were and subsequently realized how awesome they were in ASP.NET MVC also and started using them. In short, I love Rails (and Ruby in general).  I also love ASP.NET MVC and Entity Framework and yes I still love C++.  I have varying degrees of love for them individually at any given moment and it is likely to shift based on the current project I am working on.  I know you're thinking it so before you ask the question. "Which do I use when?", I'm going to give the standard developer answer of: It depends.  There are a lot of factors that I am not going to even go into that would go into a decision.  The most basic question I would ask though is,  does this project depend on .NET?  If it does, then I'd say that ASP.NET MVC is probably going to be the more logical choice and I am going to leave it at that.  I am working on projects right now in both technologies and I don't see that changing anytime soon (one project even uses both). With all that being said, you'll find me at code camps, conferences and user groups presenting on .NET, Ruby or both, writing about .NET and Ruby and I will likely be blogging on both in the future.  I know of others that have successfully branched out to other communities and with any luck I'll be successful at it too. On a (sorta) side note, I read a post by Justin Etheredge the other day that pretty much sums up my feelings about Ruby as a language.  I highly recommend checking it out: What Is So Great About Ruby?

  Read the article

• Polite busy-waiting with WRPAUSE on SPARC

  - by Dave

  Unbounded busy-waiting is an poor idea for user-space code, so we typically use spin-then-block strategies when, say, waiting for a lock to be released or some other event. If we're going to spin, even briefly, then we'd prefer to do so in a manner that minimizes performance degradation for other sibling logical processors ("strands") that share compute resources. We want to spin politely and refrain from impeding the progress and performance of other threads — ostensibly doing useful work and making progress — that run on the same core. On a SPARC T4, for instance, 8 strands will share a core, and that core has its own L1 cache and 2 pipelines. On x86 we have the PAUSE instruction, which, naively, can be thought of as a hardware "yield" operator which temporarily surrenders compute resources to threads on sibling strands. Of course this helps avoid intra-core performance interference. On the SPARC T2 our preferred busy-waiting idiom was "RD %CCR,%G0" which is a high-latency no-nop. The T4 provides a dedicated and extremely useful WRPAUSE instruction. The processor architecture manuals are the authoritative source, but briefly, WRPAUSE writes a cycle count into the the PAUSE register, which is ASR27. Barring interrupts, the processor then delays for the requested period. There's no need for the operating system to save the PAUSE register over context switches as it always resets to 0 on traps. Digressing briefly, if you use unbounded spinning then ultimately the kernel will preempt and deschedule your thread if there are other ready threads than are starving. But by using a spin-then-block strategy we can allow other ready threads to run without resorting to involuntary time-slicing, which operates on a long-ish time scale. Generally, that makes your application more responsive. In addition, by blocking voluntarily we give the operating system far more latitude regarding power management. Finally, I should note that while we have OS-level facilities like sched_yield() at our disposal, yielding almost never does what you'd want or naively expect. Returning to WRPAUSE, it's natural to ask how well it works. To help answer that question I wrote a very simple C/pthreads benchmark that launches 8 concurrent threads and binds those threads to processors 0..7. The processors are numbered geographically on the T4, so those threads will all be running on just one core. Unlike the SPARC T2, where logical CPUs 0,1,2 and 3 were assigned to the first pipeline, and CPUs 4,5,6 and 7 were assigned to the 2nd, there's no fixed mapping between CPUs and pipelines in the T4. And in some circumstances when the other 7 logical processors are idling quietly, it's possible for the remaining logical processor to leverage both pipelines. Some number T of the threads will iterate in a tight loop advancing a simple Marsaglia xor-shift pseudo-random number generator. T is a command-line argument. The main thread loops, reporting the aggregate number of PRNG steps performed collectively by those T threads in the last 10 second measurement interval. The other threads (there are 8-T of these) run in a loop busy-waiting concurrently with the T threads. We vary T between 1 and 8 threads, and report on various busy-waiting idioms. The values in the table are the aggregate number of PRNG steps completed by the set of T threads. The unit is millions of iterations per 10 seconds. For the "PRNG step" busy-waiting mode, the busy-waiting threads execute exactly the same code as the T worker threads. We can easily compute the average rate of progress for individual worker threads by dividing the aggregate score by the number of worker threads T. I should note that the PRNG steps are extremely cycle-heavy and access almost no memory, so arguably this microbenchmark is not as representative of "normal" code as it could be. And for the purposes of comparison I included a row in the table that reflects a waiting policy where the waiting threads call poll(NULL,0,1000) and block in the kernel. Obviously this isn't busy-waiting, but the data is interesting for reference. _table { border:2px black dotted; margin: auto; width: auto; } _tr { border: 2px red dashed; } _td { border: 1px green solid; } _table { border:2px black dotted; margin: auto; width: auto; } _tr { border: 2px red dashed; } td { background-color : #E0E0E0 ; text-align : right ; } th { text-align : left ; } td { background-color : #E0E0E0 ; text-align : right ; } th { text-align : left ; } Aggregate progress T = #worker threads Wait Mechanism for 8-T threadsT=1T=2T=3T=4T=5T=6T=7T=8 Park thread in poll() 32653347334833483348334833483348 no-op 415 831 124316482060249729303349 RD %ccr,%g0 "pause" 14262429269228623013316232553349 PRNG step 412 829 124616702092251029303348 WRPause(8000) 32443361333133483349334833483348 WRPause(4000) 32153308331533223347334833473348 WRPause(1000) 30853199322432513310334833483348 WRPause(500) 29173070315032223270330933483348 WRPause(250) 26942864294930773205338833483348 WRPause(100) 21552469262227902911321433303348

  Read the article

• Best Method For Evaluating Existing Software or New Software

  How many of us have been faced with having to decide on an off-the-self or a custom built component, application, or solution to integrate in to an existing system or to be the core foundation of a new system? What is the best method for evaluating existing software or new software still in the design phase? One of the industry preferred methodologies to use is the Active Reviews for Intermediate Designs (ARID) evaluation process.  ARID is a hybrid mixture of the Active Design Review (ADR) methodology and the Architectural Tradeoff Analysis Method (ATAM). So what is ARID? ARD’s main goal is to ensure quality, detailed designs in software. One way in which it does this is by empowering reviewers by assigning generic open ended survey questions. This approach attempts to remove the possibility for allowing the standard answers such as “Yes” or “No”. The ADR process ignores the “Yes”/”No” questions due to the fact that they can be leading based on how the question is asked. Additionally these questions tend to receive less thought in comparison to more open ended questions. Common Active Design Review Questions What possible exceptions can occur in this component, application, or solution? How should exceptions be handled in this component, application, or solution? Where should exceptions be handled in this component, application, or solution? How should the component, application, or solution flow based on the design? What is the maximum execution time for every component, application, or solution? What environments can this component, application, or solution? What data dependencies does this component, application, or solution have? What kind of data does this component, application, or solution require? Ok, now I know what ARID is, how can I apply? Let’s imagine that your organization is going to purchase an off-the-shelf (OTS) solution for its customer-relationship management software. What process would we use to ensure that the correct purchase is made? If we use ARID, then we will have a series of 9 steps broken up by 2 phases in order to ensure that the correct OTS solution is purchases. Phase 1 Identify the Reviewers Prepare the Design Briefing Prepare the Seed Scenarios Prepare the Materials When identifying reviewers for a design it is preferred that they be pulled from a candidate pool comprised of developers that are going to implement the design. The believe is that developers actually implementing the design will have more a vested interest in ensuring that the design is correct prior to the start of code. Design debriefing consist of a summary of the design, examples of the design solving real world examples put in to use and should be no longer than two hours typically. The primary goal of this briefing is to adequately summarize the design so that the review members could actually implement the design. In the example of purchasing an OTS product I would attempt to review my briefing prior to its distribution with the review facilitator to ensure that nothing was excluded that should have not been. This practice will also allow me to test the length of the briefing to ensure that can be delivered in an appropriate about of time. Seed Scenarios are designed to illustrate conceptualized scenarios when applied with a set of sample data. These scenarios can then be used by the reviewers in the actual evaluation of the software, All materials needed for the evaluation should be prepared ahead of time so that they can be reviewed prior to and during the meeting. Materials Included: Presentation Seed Scenarios Review Agenda Phase 2 Present ARID Present Design Brainstorm and prioritize scenarios Apply scenarios Summarize Prior to the start of any ARID review meeting the Facilitator should define the remaining steps of ARID so that all the participants know exactly what they are doing prior to the start of the review process. Once the ARID rules have been laid out, then the lead designer presents an overview of the design which typically takes about two hours. During this time no questions about the design or rational are allowed to be asked by the review panel as a standard, but they are written down for use latter in the process. After the presentation the list of compiled questions is then summarized and sent back to the lead designer as areas that need to be addressed further. In the example of purchasing an OTS product issues could arise regarding security, the implementation needed or even if this is this the correct product to solve the needed solution. After the Design presentation a brainstorming and prioritize scenarios process begins by reducing the seed scenarios down to just the highest priority scenarios.  These will then be used to test the design for suitability. Once the selected scenarios have been defined the reviewers apply the examples provided in the presentation to the scenarios. The intended output of this process is to provide code or pseudo code that makes use of the examples provided while solving the selected seed scenarios. As a standard rule, the designers of the systems are not allowed to help the review board unless they all become stuck. When this occurs it is documented and along with the reason why the designer needed to help the review panel back on track. Once all of the scenarios have been completed the review facilitator reviews with the group issues that arise during the process. Then the reviewers will be polled as to efficacy of the review experience. References: Clements, Paul., Kazman, Rick., Klien, Mark. (2002). Evaluating Software Architectures: Methods and Case Studies Indianapolis, IN: Addison-Wesley

  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

• Developing Schema Compare for Oracle (Part 5): Query Snapshots

  - by Simon Cooper

  If you've emailed us about a bug you've encountered with the EAP or beta versions of Schema Compare for Oracle, we probably asked you to send us a query snapshot of your databases. Here, I explain what a query snapshot is, and how it helps us fix your bug. Problem 1: Debugging users' bug reports When we started the Schema Compare project, we knew we were going to get problems with users' databases - configurations we hadn't considered, features that weren't installed, unicode issues, wierd dependencies... With SQL Compare, users are generally happy to send us a database backup that we can restore using a single RESTORE DATABASE command on our test servers and immediately reproduce the problem. Oracle, on the other hand, would be a lot more tricky. As Oracle generally has a 1-to-1 mapping between instances and databases, any databases users sent would have to be restored to their own instance. Furthermore, the number of steps required to get a properly working database, and the size of most oracle databases, made it infeasible to ask every customer who came across a bug during our beta program to send us their databases. We also knew that there would be lots of issues with data security that would make it hard to get backups. So we needed an easier way to be able to debug customers issues and sort out what strange schema data Oracle was returning. Problem 2: Test execution time Another issue we knew we would have to solve was the execution time of the tests we would produce for the Schema Compare engine. Our initial prototype showed that querying the data dictionary for schema information was going to be slow (at least 15 seconds per database), and this is generally proportional to the size of the database. If you're running thousands of tests on the same databases, each one registering separate schemas, not only would the tests would take hours and hours to run, but the test servers would be hammered senseless. The solution To solve these, we needed to be able to populate the schema of a database without actually connecting to it. Well, the IDataReader interface is the primary way we read data from an Oracle server. The data dictionary queries we use return their data in terms of simple strings and numbers, which we then process and reconstruct into an object model, and the results of these queries are identical for identical schemas. So, we can record the raw results of the queries once, and then replay these results to construct the same object model as many times as required without needing to actually connect to the original database. This is what query snapshots do. They are binary files containing the raw unprocessed data we get back from the oracle server for all the queries we run on the data dictionary to get schema information. The core of the query snapshot generation takes the results of the IDataReader we get from running queries on Oracle, and passes the row data to a BinaryWriter that writes it straight to a file. The query snapshot can then be replayed to create the same object model; when the results of a specific query is needed by the population code, we can simply read the binary data stored in the file on disk and present it through an IDataReader wrapper. This is far faster than querying the server over the network, and allows us to run tests in a reasonable time. They also allow us to easily debug a customers problem; using a simple snapshot generation program, users can generate a query snapshot that could be sent along with a bug report that we can immediately replay on our machines to let us debug the issue, rather than having to obtain database backups and restore databases to test systems. There are also far fewer problems with data security; query snapshots only contain schema information, which is generally less sensitive than table data. Query snapshots implementation However, actually implementing such a feature did have a couple of 'gotchas' to it. My second blog post detailed the development of the dependencies algorithm we use to ensure we get all the dependencies in the database, and that algorithm uses data from both databases to find all the needed objects - what database you're comparing to affects what objects get populated from both databases. We get information on these additional objects using an appropriate WHERE clause on all the population queries. So, in order to accurately replay the results of querying the live database, the query snapshot needs to be a snapshot of a comparison of two databases, not just populating a single database. Furthermore, although the code population queries (eg querying all_tab_cols to get column information) can simply be passed straight from the IDataReader to the BinaryWriter, we need to hook into and run the live dependencies algorithm while we're creating the snapshot to ensure we get the same WHERE clauses, and the same query results, as if we were populating straight from a live system. We also need to store the results of the dependencies queries themselves, as the resulting dependency graph is stored within the OracleDatabase object that is produced, and is later used to help order actions in synchronization scripts. This is significantly helped by the dependencies algorithm being a deterministic algorithm - given the same input, it will always return the same output. Therefore, when we're replaying a query snapshot, and processing dependency information, we simply have to return the results of the queries in the order we got them from the live database, rather than trying to calculate the contents of all_dependencies on the fly. Query snapshots are a significant feature in Schema Compare that really helps us to debug problems with the tool, as well as making our testers happier. Although not really user-visible, they are very useful to the development team to help us fix bugs in the product much faster than we otherwise would be able to.

  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

• Taking a Flying Leap

  - by Lance Shaw

  Yesterday, I went skydiving with three of my children.  It was thrilling, scary, invigorating and exciting. While there is obvious risk involved, the reward and feeling of success was well worth it. You might already be wondering what skydiving would have to with WebCenter, so let me explain. Implementing a skydiving program and becoming an instructor does not happen overnight.  It does not happen with the purchase of the needed technology. Not one of us would go out, buy a parachute, the harnesses, helmet and all the gear and be able to convince anyone that we are now ready to be a skydiving instructor. The fact is that obtaining the technology is merely a small piece of the overall process and so is the case with managing content in your company. You don't just buy the right software (Oracle WebCenter Content) and go to your boss and declare information management success. There is planning, research and effort that goes into deploying software of any kind and especially when it is as mission-critical to the success of your business as Enterprise Content Management. To become a certified skydiving instructor takes at least 3 years of commitment and often longer. In the United States, candidates must complete over 500 solo jumps of their own over a minimum of 36 months and then must complete additional rigorous training under observation.  When you consider the amount of time and effort involved, it's not unlike getting a college degree and anyone that has trusted their lives to one of these instructors will no doubt appreciate their dedication to the curriculum.  Implementing an ECM system won't take that long, but it certainly requires commitment, analysis and consideration. But guess what?  Humans are involved and that means that mistakes can happen and that rules change.  This struck me while reading an excellent post on darkreading.com by Glenn S. Phillips entitled "Mission Impossible: 4 Reasons Compliance is Impossible".  His over-arching point was that with information management and security, environments change and people are involved meaning the work is never done.  He stated that you can never claim your compliance efforts are complete because of the following reasons. People are involved.  And lets face it, some are more trustworthy than others. Change is Constant. There is always some new technology coming along that is disruptive. Consumer grade cloud file sharing and sync tools come to mind here. Compliance is interpreted, not defined.  Laws and the judges that read them are always on the move. Technology is a tool, not a complete solution. There is no magic pill. The skydiving analogy holds true here as well.  Ultimately, a single person packs your parachute.  For obvious reasons, you prefer that this person be trustworthy but there are no absolute guarantees of a 100% error-free scenario.  Weather and wind conditions are never a constant and the best-laid plans for a great day of skydiving are easily disrupted by forces outside of your control.  Rules and regulations vary by location and may be updated at any time and as I mentioned early on, even the best technology on its own will only get you started. The good news is that, like skydiving, with the right technology, the right planning, the right team and a proper understanding of the rules and regulations that govern your industry, your ECM deployment can be a great success.  Failure to plan for any of the 4 factors that Glenn outlined in his article will certainly put your deployment and maybe even your company at risk, so consider them carefully. As a final aside, for those of you who consider skydiving an incredibly dangerous and risky pastime, consider this comparative statistic.  In 2012, the U.S. Parachute Association recorded 19 fatal skydiving accidents in the U.S. out of roughly 3.1 million jumps.  That’s 0.006 fatalities per 1,000 jumps. By comparison, the U.S. National Highway Traffic Safety Administration reports that there were 34,080 deaths due to car accidents in 2012.  Based on the percentages, one could argue that it is safer to jump out of a plane than to drive to the airport where the skydiving will take place. While the way you manage, secure, classify, control, retain and dispose of company files may not carry as much risk as driving or skydiving, it certainly carries risk for the organization when not planned and deployed appropriately.  Consider all the factors involved in your organization as you make your content management plans.  For additional areas of consideration, be sure to download our free whitepaper on the topic entitled "The Top 10 Criteria for Choosing an ECM System" which is available for download here.

  Read the article

• Subterranean IL: The ThreadLocal type

  - by Simon Cooper

  I came across ThreadLocal<T> while I was researching ConcurrentBag. To look at it, it doesn't really make much sense. What's all those extra Cn classes doing in there? Why is there a GenericHolder<T,U,V,W> class? What's going on? However, digging deeper, it's a rather ingenious solution to a tricky problem. Thread statics Declaring that a variable is thread static, that is, values assigned and read from the field is specific to the thread doing the reading, is quite easy in .NET: [ThreadStatic] private static string s_ThreadStaticField; ThreadStaticAttribute is not a pseudo-custom attribute; it is compiled as a normal attribute, but the CLR has in-built magic, activated by that attribute, to redirect accesses to the field based on the executing thread's identity. TheadStaticAttribute provides a simple solution when you want to use a single field as thread-static. What if you want to create an arbitary number of thread static variables at runtime? Thread-static fields can only be declared, and are fixed, at compile time. Prior to .NET 4, you only had one solution - thread local data slots. This is a lesser-known function of Thread that has existed since .NET 1.1: LocalDataStoreSlot threadSlot = Thread.AllocateNamedDataSlot("slot1"); string value = "foo"; Thread.SetData(threadSlot, value); string gettedValue = (string)Thread.GetData(threadSlot); Each instance of LocalStoreDataSlot mediates access to a single slot, and each slot acts like a separate thread-static field. As you can see, using thread data slots is quite cumbersome. You need to keep track of LocalDataStoreSlot objects, it's not obvious how instances of LocalDataStoreSlot correspond to individual thread-static variables, and it's not type safe. It's also relatively slow and complicated; the internal implementation consists of a whole series of classes hanging off a single thread-static field in Thread itself, using various arrays, lists, and locks for synchronization. ThreadLocal<T> is far simpler and easier to use. ThreadLocal ThreadLocal provides an abstraction around thread-static fields that allows it to be used just like any other class; it can be used as a replacement for a thread-static field, it can be used in a List<ThreadLocal<T>>, you can create as many as you need at runtime. So what does it do? It can't just have an instance-specific thread-static field, because thread-static fields have to be declared as static, and so shared between all instances of the declaring type. There's something else going on here. The values stored in instances of ThreadLocal<T> are stored in instantiations of the GenericHolder<T,U,V,W> class, which contains a single ThreadStatic field (s_value) to store the actual value. This class is then instantiated with various combinations of the Cn types for generic arguments. In .NET, each separate instantiation of a generic type has its own static state. For example, GenericHolder<int,C0,C1,C2> has a completely separate s_value field to GenericHolder<int,C1,C14,C1>. This feature is (ab)used by ThreadLocal to emulate instance thread-static fields. Every time an instance of ThreadLocal is constructed, it is assigned a unique number from the static s_currentTypeId field using Interlocked.Increment, in the FindNextTypeIndex method. The hexadecimal representation of that number then defines the specific Cn types that instantiates the GenericHolder class. That instantiation is therefore 'owned' by that instance of ThreadLocal. This gives each instance of ThreadLocal its own ThreadStatic field through a specific unique instantiation of the GenericHolder class. Although GenericHolder has four type variables, the first one is always instantiated to the type stored in the ThreadLocal<T>. This gives three free type variables, each of which can be instantiated to one of 16 types (C0 to C15). This puts an upper limit of 4096 (163) on the number of ThreadLocal<T> instances that can be created for each value of T. That is, there can be a maximum of 4096 instances of ThreadLocal<string>, and separately a maximum of 4096 instances of ThreadLocal<object>, etc. However, there is an upper limit of 16384 enforced on the total number of ThreadLocal instances in the AppDomain. This is to stop too much memory being used by thousands of instantiations of GenericHolder<T,U,V,W>, as once a type is loaded into an AppDomain it cannot be unloaded, and will continue to sit there taking up memory until the AppDomain is unloaded. The total number of ThreadLocal instances created is tracked by the ThreadLocalGlobalCounter class. So what happens when either limit is reached? Firstly, to try and stop this limit being reached, it recycles GenericHolder type indexes of ThreadLocal instances that get disposed using the s_availableIndices concurrent stack. This allows GenericHolder instantiations of disposed ThreadLocal instances to be re-used. But if there aren't any available instantiations, then ThreadLocal falls back on a standard thread local slot using TLSHolder. This makes it very important to dispose of your ThreadLocal instances if you'll be using lots of them, so the type instantiations can be recycled. The previous way of creating arbitary thread-static variables, thread data slots, was slow, clunky, and hard to use. In comparison, ThreadLocal can be used just like any other type, and each instance appears from the outside to be a non-static thread-static variable. It does this by using the CLR type system to assign each instance of ThreadLocal its own instantiated type containing a thread-static field, and so delegating a lot of the bookkeeping that thread data slots had to do to the CLR type system itself! That's a very clever use of the CLR type system.

  Read the article

• Null Values And The T-SQL IN Operator

  - by Jesse

  I came across some unexpected behavior while troubleshooting a failing test the other day that took me long enough to figure out that I thought it was worth sharing here. I finally traced the failing test back to a SELECT statement in a stored procedure that was using the IN t-sql operator to exclude a certain set of values. Here’s a very simple example table to illustrate the issue: Customers CustomerId INT, NOT NULL, Primary Key CustomerName nvarchar(100) NOT NULL SalesRegionId INT NULL   The ‘SalesRegionId’ column contains a number representing the sales region that the customer belongs to. This column is nullable because new customers get created all the time but assigning them to sales regions is a process that is handled by a regional manager on a periodic basis. For the purposes of this example, the Customers table currently has the following rows: CustomerId CustomerName SalesRegionId 1 Customer A 1 2 Customer B NULL 3 Customer C 4 4 Customer D 2 5 Customer E 3   How could we write a query against this table for all customers that are NOT in sales regions 2 or 4? You might try something like this: 1: SELECT 2: CustomerId, 3: CustomerName, 4: SalesRegionId 5: FROM Customers 6: WHERE SalesRegionId NOT IN (2,4)   Will this work? In short, no; at least not in the way that you might expect. Here’s what this query will return given the example data we’re working with: CustomerId CustomerName SalesRegionId 1 Customer A 1 5 Customer E 5   I was expecting that this query would also return ‘Customer B’, since that customer has a NULL SalesRegionId. In my mind, having a customer with no sales region should be included in a set of customers that are not in sales regions 2 or 4.When I first started troubleshooting my issue I made note of the fact that this query should probably be re-written without the NOT IN clause, but I didn’t suspect that the NOT IN clause was actually the source of the issue. This particular query was only one minor piece in a much larger process that was being exercised via an automated integration test and I simply made a poor assumption that the NOT IN would work the way that I thought it should. So why doesn’t this work the way that I thought it should? From the MSDN documentation on the t-sql IN operator: If the value of test_expression is equal to any value returned by subquery or is equal to any expression from the comma-separated list, the result value is TRUE; otherwise, the result value is FALSE. Using NOT IN negates the subquery value or expression. The key phrase out of that quote is, “… is equal to any expression from the comma-separated list…”. The NULL SalesRegionId isn’t included in the NOT IN because of how NULL values are handled in equality comparisons. From the MSDN documentation on ANSI_NULLS: The SQL-92 standard requires that an equals (=) or not equal to (<>) comparison against a null value evaluates to FALSE. When SET ANSI_NULLS is ON, a SELECT statement using WHERE column_name = NULL returns zero rows even if there are null values in column_name. A SELECT statement using WHERE column_name <> NULL returns zero rows even if there are nonnull values in column_name. In fact, the MSDN documentation on the IN operator includes the following blurb about using NULL values in IN sub-queries or expressions that are used with the IN operator: Any null values returned by subquery or expression that are compared to test_expression using IN or NOT IN return UNKNOWN. Using null values in together with IN or NOT IN can produce unexpected results. If I were to include a ‘SET ANSI_NULLS OFF’ command right above my SELECT statement I would get ‘Customer B’ returned in the results, but that’s definitely not the right way to deal with this. We could re-write the query to explicitly include the NULL value in the WHERE clause: 1: SELECT 2: CustomerId, 3: CustomerName, 4: SalesRegionId 5: FROM Customers 6: WHERE (SalesRegionId NOT IN (2,4) OR SalesRegionId IS NULL)   This query works and properly includes ‘Customer B’ in the results, but I ultimately opted to re-write the query using a LEFT OUTER JOIN against a table variable containing all of the values that I wanted to exclude because, in my case, there could potentially be several hundred values to be excluded. If we were to apply the same refactoring to our simple sales region example we’d end up with: 1: DECLARE @regionsToIgnore TABLE (IgnoredRegionId INT) 2: INSERT @regionsToIgnore values (2),(4) 3:  4: SELECT 5: c.CustomerId, 6: c.CustomerName, 7: c.SalesRegionId 8: FROM Customers c 9: LEFT OUTER JOIN @regionsToIgnore r ON r.IgnoredRegionId = c.SalesRegionId 10: WHERE r.IgnoredRegionId IS NULL By performing a LEFT OUTER JOIN from Customers to the @regionsToIgnore table variable we can simply exclude any rows where the IgnoredRegionId is null, as those represent customers that DO NOT appear in the ignored regions list. This approach will likely perform better if the number of sales regions to ignore gets very large and it also will correctly include any customers that do not yet have a sales region.

  Read the article

• Documentation Changes in Solaris 11.1

  - by alanc

  One of the first places you can see Solaris 11.1 changes are in the docs, which have now been posted in the Solaris 11.1 Library on docs.oracle.com. I spent a good deal of time reviewing documentation for this release, and thought some would be interesting to blog about, but didn't review all the changes (not by a long shot), and am not going to cover all the changes here, so there's plenty left for you to discover on your own. Just comparing the Solaris 11.1 Library list of docs against the Solaris 11 list will show a lot of reorganization and refactoring of the doc set, especially in the system administration guides. Hopefully the new break down will make it easier to get straight to the sections you need when a task is at hand. Packaging System Unfortunately, the excellent in-depth guide for how to build packages for the new Image Packaging System (IPS) in Solaris 11 wasn't done in time to make the initial Solaris 11 doc set. An interim version was published shortly after release, in PDF form on the OTN IPS page. For Solaris 11.1 it was included in the doc set, as Packaging and Delivering Software With the Image Packaging System in Oracle Solaris 11.1, so should be easier to find, and easier to share links to specific pages the HTML version. Beyond just how to build a package, it includes details on how Solaris is packaged, and how package updates work, which may be useful to all system administrators who deal with Solaris 11 upgrades & installations. The Adding and Updating Oracle Solaris 11.1 Software Packages was also extended, including new sections on Relaxing Version Constraints Specified by Incorporations and Locking Packages to a Specified Version that may be of interest to those who want to keep the Solaris 11 versions of certain packages when they upgrade, such as the couple of packages that had functionality removed by an (unusual for an update release) End of Feature process in the 11.1 release. Also added in this release is a document containing the lists of all the packages in each of the major package groups in Solaris 11.1 (solaris-desktop, solaris-large-server, and solaris-small-server). While you can simply get the contents of those groups from the package repository, either via the web interface or the pkg command line, the documentation puts them in handy tables for easier side-by-side comparison, or viewing the lists before you've installed the system to pick which one you want to initially install. X Window System We've not had good X11 coverage in the online Solaris docs in a while, mostly relying on the man pages, and upstream X.Org docs. In this release, we've integrated some X coverage into the Solaris 11.1 Desktop Adminstrator's Guide, including sections on installing fonts for fontconfig or legacy X11 clients, X server configuration, and setting up remote access via X11 or VNC. Of course we continue to work on improving the docs, including a lot of contributions to the upstream docs all OS'es share (more about that another time). Security One of the things Oracle likes to do for its products is to publish security guides for administrators & developers to know how to build systems that meet their security needs. For Solaris, we started this with Solaris 11, providing a guide for sysadmins to find where the security relevant configuration options were documented. The Solaris 11.1 Security Guidelines extend this to cover new security features, such as Address Space Layout Randomization (ASLR) and Read-Only Zones, as well as adding additional guidelines for existing features, such as how to limit the size of tmpfs filesystems, to avoid users driving the system into swap thrashing situations. For developers, the corresponding document is the Developer's Guide to Oracle Solaris 11 Security, which has been the source for years for documentation of security-relevant Solaris API's such as PAM, GSS-API, and the Solaris Cryptographic Framework. For Solaris 11.1, a new appendix was added to start providing Secure Coding Guidelines for Developers, leveraging the CERT Secure Coding Standards and OWASP guidelines to provide the base recommendations for common programming languages and their standard API's. Solaris specific secure programming guidance was added via links to other documentation in the product doc set. In parallel, we updated the Solaris C Libary Functions security considerations list with details of Solaris 11 enhancements such as FD_CLOEXEC flags, additional *at() functions, and new stdio functions such as asprintf() and getline(). A number of code examples throughout the Solaris 11.1 doc set were updated to follow these recommendations, changing unbounded strcpy() calls to strlcpy(), sprintf() to snprintf(), etc. so that developers following our examples start out with safer code. The Writing Device Drivers guide even had the appendix updated to list which of these utility functions, like snprintf() and strlcpy(), are now available via the Kernel DDI. Little Things Of course all the big new features got documented, and some major efforts were put into refactoring and renovation, but there were also a lot of smaller things that got fixed as well in the nearly a year between the Solaris 11 and 11.1 doc releases - again too many to list here, but a random sampling of the ones I know about & found interesting or useful: The Privileges section of the DTrace Guide now gives users a pointer to find out how to set up DTrace privileges for non-global zones and what limitations are in place there. A new section on Recommended iSCSI Configuration Practices was added to the iSCSI configuration section when it moved into the SAN Configuration and Multipathing administration guide. The Managing System Power Services section contains an expanded explanation of the various tunables for power management in Solaris 11.1. The sample dcmd sources in /usr/demo/mdb were updated to include ::help output, so that developers like myself who follow the examples don't forget to include it (until a helpful code reviewer pointed it out while reviewing the mdb module changes for Xorg 1.12). The README file in that directory was updated to show the correct paths for installing both kernel & userspace modules, including the 64-bit variants.

  Read the article

• CS, SE, HCI, Information Science, Please recommendation for further education of the former performing art manager seeking career in IT industries? [on hold]

  - by Baek Seungjoo

  IT specialists there J Thank you very much for your collective efforts here, and I got huge help reading your professional comments and advices on each questions I have searched so far! This time, I would like to ask for your practical advices or recommendation on what I am struggling on at this moment. I am currently seeking higher education for my career transition from performing art manager and director to “IT software and/or service development and management specialist”. However, as this field is quite new to me, and there are lots of different work positions, I have no idea which grad major I better pursue in order to get qualification. Of course I know this question could sounds wired as it is kind of personal choice. But my lack of understanding on how IT software companies work in general, your practical and experience-based advice will be great help to me, who spent more than two months of self-research on net. OK. Before my question, here is my plan and history, which are quite different from those currently in IT industry I think… 1) Target Firstly, get career transition into IT service or products companies and get experiences. Eventually, pursue IT entrepreneurship in combination with my arts and cultural production and business expertise. 2) Background Career: performing arts director and manager in theatre-based scale opera and musical Art education in youth BA in literature and Chinese studies (Art & Humanities) MA in Cultural & Creative Industries (Art & Humanities) – dissertation with focus on digital prosumption and the lived experience of the prosumer. (a qualitative research on the agents in the digital world) 2) Personally Huge interest in IT hardware and software, and their trend. Skills to build up, repair, tune PCs -of course this is no more than personal hobby, but shows my interests in this field. 4) Problem Encounter a question “So, what do you think you can contribute practically in this position”. This question turn me down everytime I go through job interviews, and I decided more education in the relevant area. Here are my questions. 1) In terms of work positions in IT software companies, I wonder if I can put the comparison of what “Artists” is to “Arts Manager or Director” is what “Developer” is to “Product Manager”. (Of course, this stereotypical division of Artist-Art Manager is out of sense because the domain overlaps to some extent, and is blurring at least in my field, and they are in different contexts, but just speaking easily.) Normally, artist comes with special arts educations, and they live in their own world of artistic inspiration and creation, and they feel alive in practice and on stages. Meanwhile, from the point of staging and managing productions, the role of art manager is critical as well. Our role cares how the production appeals to the audience in effective way, how to make profit and future sustainable management through that, how to set up future strategy in consideration of the external conditions such as political and social circumstances, audience trend and level, other production trends from on-going and historical perspectives, how and what the production make voice to the society from political, economic, humanitarian stances. So, we need keen eyes on economic, political, and societal environment, have to understand human-being and their desires, must know how to make presentation and attract investors, must have sense in managing and fighting over the limited financial resource, how to extend networking and so on. It is common that the two agents create productions in collaboration (normally not in that ideal way but in conflict and fight though J ). So, we need to know each other’s expertise to some extent, for better production. What are the work positions in IT software industries equivalent to the role of “art manager” in performing arts? From my view, considering developers come with special education in the world of computer science, software engineering, or others (self-education sometimes), and they express themselves with the arts of coding, computer languages on the black screen, and make sort of their artistic production online to the audience, I guess there might be someone who collaborate with developers in creating, managing, and launching IT services or products. 2) Which education among CS, SE, HCI, Information Science, is needed for those seeking such work position? Especially for person like me. (At this moment, Information Science has the highest possibility to get in, since I lack Calculus and Math in undergrad educaiton. But please let me know irrespective of this concern, I think there are ways to back it up if CS or SE education needed in my case) 3) Which field between Information Science and HCI can be more practical background regarding job hungting? And which of them have more demands in job market? AS I checked, HCI is more close to CS than IS in its focus of study area. Thank you very much for your patience reading such a long inquiry, and I appreciate to your efforts in advance. Have a nice day in this beautiful summer.

  Read the article

• k-d tree implementation [closed]

  - by user466441

  when i run my code and debugged,i got this error - this 0x00093584 {_Myproxy=0x00000000 _Mynextiter=0x00000000 } std::_Iterator_base12 * const - _Myproxy 0x00000000 {_Mycont=??? _Myfirstiter=??? } std::_Container_proxy * _Mycont CXX0017: Error: symbol "" not found _Myfirstiter CXX0030: Error: expression cannot be evaluated + _Mynextiter 0x00000000 {_Myproxy=??? _Mynextiter=??? } std::_Iterator_base12 * but i dont know what does it means,code is this #include<iostream> #include<vector> #include<algorithm> using namespace std; struct point { float x,y; }; vector<point>pointleft(4); vector<point>pointright(4); //we are going to implement two comparison function for x and y coordinates,we need it in calculation of median (we should sort vector //by x or y according to depth informaton,is depth even or odd. bool sortby_X(point &a,point &b) { return a.x<b.x; } bool sortby_Y(point &a,point &b) { return a.y<b.y; } //so i am going to implement to median finding algorithm,one for finding median by x and another find median by y point medianx(vector<point>points) { point temp; sort(points.begin(),points.end(),sortby_X); temp=points[(points.size()/2)]; return temp; } point mediany(vector<point>points) { point temp; sort(points.begin(),points.end(),sortby_Y); temp=points[(points.size()/2)]; return temp; } //now construct basic tree structure struct Tree { float x,y; Tree(point a) { x=a.x; y=a.y; } Tree *left; Tree *right; }; Tree * build_kd( Tree *root,vector<point>points,int depth) { point temp; if(points.size()==1)// that point is as a leaf { if(root==NULL) root=new Tree(points[0]); return root; } if(depth%2==0) { temp=medianx(points); root=new Tree(temp); for(int i=0;i<points.size();i++) { if (points[i].x<temp.x) pointleft[i]=points[i]; else pointright[i]=points[i]; } } else { temp=mediany(points); root=new Tree(temp); for(int i=0;i<points.size();i++) { if(points[i].y<temp.y) pointleft[i]=points[i]; else pointright[i]=points[i]; } } return build_kd(root->left,pointleft,depth+1); return build_kd(root->right,pointright,depth+1); } void print(Tree *root) { while(root!=NULL) { cout<<root->x<<" " <<root->y; print(root->left); print(root->right); } } int main() { int depth=0; Tree *root=NULL; vector<point>points(4); float x,y; int n=4; for(int i=0;i<n;i++) { cin>>x>>y; points[i].x=x; points[i].y=y; } root=build_kd(root,points,depth); print(root); return 0; } i am trying ti implement in c++ this pseudo code tuple function build_kd_tree(int depth, set points): if points contains only one point: return that point as a leaf. if depth is even: Calculate the median x-value. Create a set of points (pointsLeft) that have x-values less than the median. Create a set of points (pointsRight) that have x-values greater than or equal to the median. else: Calculate the median y-value. Create a set of points (pointsLeft) that have y-values less than the median. Create a set of points (pointsRight) that have y-values greater than or equal to the median. treeLeft = build_kd_tree(depth + 1, pointsLeft) treeRight = build_kd_tree(depth + 1, pointsRight) return(median, treeLeft, treeRight) please help me what this error means?

  Read the article

• What are the industry metrics for average spend on dev hardware and software? [on hold]

  - by RationalGeek

  I'm trying to budget for my dev shop and compare our budget items to industry expectations. I'm hoping to find some information on what percentage of a dev's salary is generally spent on tooling, both hardware and software. Where can I find such information? If instead there is a source that looks at raw dollars that is useful, too. I can extrapolate what I need from that. NOTE: Your anecdotal evidence from your own job will not be very helpful. I'm looking for industry average statistics from a credible source. EDIT: I'm reluctant to even keep this question going based on the passionate negative responses of commenters, but I do think this is valuable information (assuming anyone will care to answer) so let me make one attempt to clarify why I'm looking for this information, and then leave it at that. I'm not sure why understanding and validating my motives is a necessary step to providing the information, but apparently that is the case, so I will do my best. Firstly, let me respond to the idea that us "management types" shouldn't use these types of metrics to evaluate budgets. I agree in part. Ideally, you should spend whatever is necessary on developers in order to keep them fully happy and productive. And this is true of all employees. However, companies operate in a world of limited resources, and every dollar spent in one area means a dollar not spent in another. So it is not enough to simply say "I need to spend $10,000 per developer next year" without having some way to justify that position. One way to help justify it is to compare yourself against the industry. If it is the case that on average a software shops spends 5% (making up that number) of their total development budget (salaries being the large portion of the other 95%, for arguments sake), and I'm only spending 3%, it helps in the justification process. So, it is not my intent to use this information to limit what I spend on developers, but rather to arm myself with the necessary justification to spend what I need to spend on developers to give them the best tools I can. I have been a developer for many years and I understand the need for proper tooling. Next, let's examine the idea that even considering the relationship between a spend on developer salaries and developer tooling is ludicrous and should be banned from budgetary thinking. As Jimmy Hoffa put it in their comment, it's like saying "I'm going to spend no more than 10% of median employee salary on light bulbs and coffee from now on.". Well, yes, it is like saying that, and from a budgeting perspective, this is a useful way to look at things. If you know that, on average, an employee consumes X dollars of coffee a year, then you can project a coffee budget based on that. And you can compare it to an industry metric to understand where you fall: do you spend more on coffee than other companies or less? Why might this be? If you are a coffee supply manager, that seems like a useful thought process. The same seems to hold true for developers. Now, on to the idea that I need to compare "apples to apples" and only look at other shops that are in the same place geographically, the same business, the same application architecture, and the same development frameworks. I guess if I could find such a statistic that said "a shop that is exactly identical to yours spends X on developer tooling" it would be wonderful. But there is plenty of value in an average statistic. Here's an analogy: let's say you are working on a household budget and need to decide how much to spend on groceries. Is it enough to know that the average consumer spends 15% on groceries and therefore decide that you will budget exactly 15%? No. You have to tweak your budget based on your individual needs and situation. But the generalized statistic does help in this evaluation. You can know if your budget is grossly off from what others are doing, and this can help you figure out why this is. So, I will concede the point that it would be better to find statistics that align to my shop, though I think any statistics I could find would be useful for what I'm doing. In that light, let's say that my shop is mostly focused on ASP.NET web applications. That doesn't map perfectly to reality because large enterprises have very heterogenous IT environments. But if I was going to pick one technology that is our focus that would be it. But, if you were to point me at some statistics that are related to a Linux shop doing embedded Java applications, I would still find it useful as a point of comparison. SUMMARY: Let me try to rephrase my question. I'm trying to find industry metrics on how much dev shops spend on developer tooling, both hardware and software. I don't so much care whether it is expressed as a percentage of total budget or as X dollars per dev or as Y percentage of salary. Any metric would be useful. If there are metrics that are specific to ASP.NET dev shops in the Northeast US, all the better, but I would be happy to find anything.

  Read the article

• Solaris 11.1 changes building of code past the point of __NORETURN

  - by alanc

  While Solaris 11.1 was under development, we started seeing some errors in the builds of the upstream X.Org git master sources, such as: "Display.c", line 65: Function has no return statement : x_io_error_handler "hostx.c", line 341: Function has no return statement : x_io_error_handler from functions that were defined to match a specific callback definition that declared them as returning an int if they did return, but these were calling exit() instead of returning so hadn't listed a return value. These had been generating warnings for years which we'd been ignoring, but X.Org has made enough progress in cleaning up code for compiler warnings and static analysis issues lately, that the community turned up the default error levels, including the gcc flag -Werror=return-type and the equivalent Solaris Studio cc flags -v -errwarn=E_FUNC_HAS_NO_RETURN_STMT, so now these became errors that stopped the build. Yet on Solaris, gcc built this code fine, while Studio errored out. Investigation showed this was due to the Solaris headers, which during Solaris 10 development added a number of annotations to the headers when gcc was being used for the amd64 kernel bringup before the Studio amd64 port was ready. Since Studio did not support the inline form of these annotations at the time, but instead used #pragma for them, the definitions were only present for gcc. To resolve this, I fixed both sides of the problem, so that it would work for building new X.Org sources on older Solaris releases or with older Studio compilers, as well as fixing the general problem before it broke more software building on Solaris. To the X.Org sources, I added the traditional Studio #pragma does_not_return to recognize that functions like exit() don't ever return, in patches such as this Xserver patch. Adding a dummy return statement was ruled out as that introduced unreachable code errors from compilers and analyzers that correctly realized you couldn't reach that code after a return statement. And on the Solaris 11.1 side, I updated the annotation definitions in <sys/ccompile.h> to enable for Studio 12.0 and later compilers the annotations already existing in a number of system headers for functions like exit() and abort(). If you look in that file you'll see the annotations we currently use, though the forms there haven't gone through review to become a Committed interface, so may change in the future. Actually getting this integrated into Solaris though took a bit more work than just editing one header file. Our ELF binary build comparison tool, wsdiff, actually showed a large number of differences in the resulting binaries due to the compiler using this information for branch prediction, code path analysis, and other possible optimizations, so after comparing enough of the disassembly output to be comfortable with the changes, we also made sure to get this in early enough in the release cycle so that it would get plenty of test exposure before the release. It also required updating quite a bit of code to avoid introducing new lint or compiler warnings or errors, and people building applications on top of Solaris 11.1 and later may need to make similar changes if they want to keep their build logs similarly clean. Previously, if you had a function that was declared with a non-void return type, lint and cc would warn if you didn't return a value, even if you called a function like exit() or panic() that ended execution. For instance: #include <stdlib.h> int callback(int status) { if (status == 0) return status; exit(status); } would previously require a never executed return 0; after the exit() to avoid lint warning "function falls off bottom without returning value". Now the compiler & lint will both issue "statement not reached" warnings for a return 0; after the final exit(), allowing (or in some cases, requiring) it to be removed. However, if there is no return statement anywhere in the function, lint will warn that you've declared a function returning a value that never does so, suggesting you can declare it as void. Unfortunately, if your function signature is required to match a certain form, such as in a callback, you not be able to do so, and will need to add a /* LINTED */ to the end of the function. If you need your code to build on both a newer and an older release, then you will either need to #ifdef these unreachable statements, or, to keep your sources common across releases, add to your sources the corresponding #pragma recognized by both current and older compiler versions, such as: #pragma does_not_return(exit) #pragma does_not_return(panic) Hopefully this little extra work is paid for by the compilers & code analyzers being able to better understand your code paths, giving you better optimizations and more accurate errors & warning messages.

  Read the article

• Using DisplayTag library, I want to have the currently selected row have a unique custom class using

  - by Mary

  I have been trying to figure out how to highlight the selected row in a table. In my jsp I have jsp scriplet that can get access to the id of the row the displaytag library is creating. I want to compare it to the the id of the current row selected by the user ${currentNoteId}. Right now if the row id = 849 (hardcoded) the class "currentClass" is added to just that row of the table. I need to change the 849 for the {$currentNoteId} and I don't know how to do it. I am using java, Spring MVC. The jsp: ... <% request.setAttribute("dyndecorator", new org.displaytag.decorator.TableDecorator() { public String addRowClass() { edu.ilstu.ais.advisorApps.business.Note row = (edu.ilstu.ais.advisorApps.business.Note)getCurrentRowObject(); String rowId = row.getId(); if ( rowId.equals("849") ) { return "currentClass"; } return null; } }); %> <c:set var="currentNoteId" value="${studentNotes.currentNote.id}"/> ... <display:table id="noteTable" name="${ studentNotes.studentList }" pagesize="20" requestURI="notesView.form.html" decorator="dyndecorator"> <display:column title="Select" class="yui-button-match" href="/notesView.form.html" paramId="note.id" paramProperty="id"> <input type="button" class="yui-button-match2" name="select" value="Select"/> </display:column> <display:column property="userName" title="Created By" sortable="true"/> <display:column property="createDate" title="Created On" sortable="true" format="{0,date,MM/dd/yy hh:mm:ss a}"/> <display:column property="detail" title="Detail" sortable="true"/> </display:table> ... This could also get done using javascript and that might be best, but the documentation suggested this so I thought I would try it. I cannot find an example anywhere using the addRowClass() unless the comparison is to a field already in the row (a dollar amount is used in the documentation example) or hardcoded in like the "849" id. Thanks for any help you can provide.

  Read the article

< Previous Page | 68 69 70 71 72 73 74 75 76 77 78 79  | Next Page >