Search Results

Search found 666 results on 27 pages for 'profiler'.

Page 7/27 | < Previous Page | 3 4 5 6 7 8 9 10 11 12 13 14  | Next Page >

  • How to profile my C++ application on linux

    - by richard
    HI, I would like to profile my c++ application on linux. I would like to find out how much time my application spent on CPU processing vs time spent on block by IO/being idle. I know there is a profile tool call valgrind on linux. But it breaks down time spent on each method, and it does not give me an overall picture of how much time spent on CPU processing vs idle? Or is there a way to do that with valgrind. Thank you.

    Read the article

  • Profiling With Visual Studio Team System

    - by Rotem
    Hi, I'm using Visual Studio Team System 2008 to run Load Tests. I have a test that executes a web service request and I would like to know how much time was spent in each layer of my application e.g. Time spent in IIS, Time spent in my Server application Time spent in SQL Server Can I get this sort of information by setting the performance counters in my load test properly? Thanks

    Read the article

  • How to Get the Method/Function Call Trace for a Specific Run?

    - by JackWM
    Given a Java or JavaScript program, after its execution, print out a sequence of calls. The calls are in invocation order. E.g. main() { A(); } A() { B(); C(); } Then the call trace should be: main -> A() -> B() -> C() Is there any tool that can profile and output this kind of information? It seems this is common a need for debugging or performance tuning. I noticed that some profilers can do this, but I prefer a simpler/easy-to-use one. Thanks!

    Read the article

  • How do I display java.lang.* object allocations in Eclipse profiler?

    - by Martin Wickman
    I am profiling an application using the Eclipse profiler. I am particularly interested in number of allocated object instances of classes from java.lang (for instance java.lang.String or java.util.HashMap). I also want to know stuff like number of calls to String.equals() etc. I use the "Object Allocations" tab and I shows all classes in my application and a count. It also shows all int[], byte[], long[] etc, but there is no mention of any standard java classes. For instance, this silly code: public static void main(String[] args) { Object obj[] = new Object[1000]; for (int i = 0; i < 1000; i++) { obj[i] = new StringBuffer("foo" + i); } System.out.println (obj[30]); } Shows up in the Object Allocations tab as 7 byte[]s, 4 char[]s and 2 int[]s. It doesn't matter if I use 1000 or 1 iterations. It seems the profiler simply ignores everything that is in any of the java.* packages. The same applies to Execution Statistics as well. Any idea how to display instances of java.* in the Eclipse Profiler?

    Read the article

  • How do I get SQL Profiler to show statements with column names like 'password'?

    - by Kev
    I'm profiling a database just now and need to see the UPDATE and INSERT statements being executed on a particular table. However, because the table has a 'Password' column the SQL Profiler is being understandingly cautious and replacing the TextData column with: -- 'password' was found in the text of this event. -- The text has been replaced with this comment for security reasons. How do I prevent it doing this because I need to see the SQL statement being executed?

    Read the article

  • Using the JRockit Flight Recorder as an Exception Profiler.

    - by Marcus Hirt
    There is a lot of new data points in the JRockit Flight Recorder compared to the data available in the old JRA. One set of data deals with exceptions and where they are thrown. In JRA, it was possible to tell how many exceptions were thrown, but it was not possible to determine from where they were thrown. Here is how to do a recording with exception profiling enabled from JRockit Mission Control. 1. Right click on the JVM to profile, select Start Flight Recording. 2. Select the Profiling with Exceptions template.   3. Wait for the recording to finish. The count down for the time left will show in the Flight Recorder Control view. 4. When done the recording will automatically be downloaded and displayed. To show the exceptions, go to the Code | Exceptions tab.

    Read the article

  • java System.nanoTime is really slow. Is it possible to implement a high performance java profiler?

    - by willpowerforever
    I did a test and found the overhead of a function call to System.nanoTime() is at least 500 ns on my machine. Seemed that it is very hard to have a high performance java profiler. For enterprise software, suppose a function takes about 350 seconds and has 12,500,000,000 times of method calls. Therefore, the number of calls to System.nanoTime() is: 12,500,000,000 * 2 = 25,000,000,000 (one for start timestamp, one for end timestamp) And the overhead of System.nanoTime in total is: 500 ns * 25,000,000,000 = 500 * 25000 s = 12500000s. Note: all data from real case. Any better way to acquire the timestamp?

    Read the article

  • How can I create a profiler GUI in WPF?

    - by thehan
    I am looking to create a task profiler in WPF with an UI similar to the one of incredibuild. Eg, timed tasks appear as strips on their respective line. Have a look at: http://baoz.net/wp-content/2009/06/ib1.png to see what I mean. I am unsure what the best way to tackle this problem is? Should I override the Panel class creating a custom layout panel for this or might there be an existing control/library out there that can fit my needs. I have so far had a look at various charting controls but am unsure of weither they can be tweaked to achieve this behaviour or not. I am really hoping for someone who has been using WPF for a while to help me narrow down my search. It is not exactly easy to know what to google for either :)

    Read the article

  • How to find and fix performance problems in ORM powered applications

    - by FransBouma
    Once in a while we get requests about how to fix performance problems with our framework. As it comes down to following the same steps and looking into the same things every single time, I decided to write a blogpost about it instead, so more people can learn from this and solve performance problems in their O/R mapper powered applications. In some parts it's focused on LLBLGen Pro but it's also usable for other O/R mapping frameworks, as the vast majority of performance problems in O/R mapper powered applications are not specific for a certain O/R mapper framework. Too often, the developer looks at the wrong part of the application, trying to fix what isn't a problem in that part, and getting frustrated that 'things are so slow with <insert your favorite framework X here>'. I'm in the O/R mapper business for a long time now (almost 10 years, full time) and as it's a small world, we O/R mapper developers know almost all tricks to pull off by now: we all know what to do to make task ABC faster and what compromises (because there are almost always compromises) to deal with if we decide to make ABC faster that way. Some O/R mapper frameworks are faster in X, others in Y, but you can be sure the difference is mainly a result of a compromise some developers are willing to deal with and others aren't. That's why the O/R mapper frameworks on the market today are different in many ways, even though they all fetch and save entities from and to a database. I'm not suggesting there's no room for improvement in today's O/R mapper frameworks, there always is, but it's not a matter of 'the slowness of the application is caused by the O/R mapper' anymore. Perhaps query generation can be optimized a bit here, row materialization can be optimized a bit there, but it's mainly coming down to milliseconds. Still worth it if you're a framework developer, but it's not much compared to the time spend inside databases and in user code: if a complete fetch takes 40ms or 50ms (from call to entity object collection), it won't make a difference for your application as that 10ms difference won't be noticed. That's why it's very important to find the real locations of the problems so developers can fix them properly and don't get frustrated because their quest to get a fast, performing application failed. Performance tuning basics and rules Finding and fixing performance problems in any application is a strict procedure with four prescribed steps: isolate, analyze, interpret and fix, in that order. It's key that you don't skip a step nor make assumptions: these steps help you find the reason of a problem which seems to be there, and how to fix it or leave it as-is. Skipping a step, or when you assume things will be bad/slow without doing analysis will lead to the path of premature optimization and won't actually solve your problems, only create new ones. The most important rule of finding and fixing performance problems in software is that you have to understand what 'performance problem' actually means. Most developers will say "when a piece of software / code is slow, you have a performance problem". But is that actually the case? If I write a Linq query which will aggregate, group and sort 5 million rows from several tables to produce a resultset of 10 rows, it might take more than a couple of milliseconds before that resultset is ready to be consumed by other logic. If I solely look at the Linq query, the code consuming the resultset of the 10 rows and then look at the time it takes to complete the whole procedure, it will appear to me to be slow: all that time taken to produce and consume 10 rows? But if you look closer, if you analyze and interpret the situation, you'll see it does a tremendous amount of work, and in that light it might even be extremely fast. With every performance problem you encounter, always do realize that what you're trying to solve is perhaps not a technical problem at all, but a perception problem. The second most important rule you have to understand is based on the old saying "Penny wise, Pound Foolish": the part which takes e.g. 5% of the total time T for a given task isn't worth optimizing if you have another part which takes a much larger part of the total time T for that same given task. Optimizing parts which are relatively insignificant for the total time taken is not going to bring you better results overall, even if you totally optimize that part away. This is the core reason why analysis of the complete set of application parts which participate in a given task is key to being successful in solving performance problems: No analysis -> no problem -> no solution. One warning up front: hunting for performance will always include making compromises. Fast software can be made maintainable, but if you want to squeeze as much performance out of your software, you will inevitably be faced with the dilemma of compromising one or more from the group {readability, maintainability, features} for the extra performance you think you'll gain. It's then up to you to decide whether it's worth it. In almost all cases it's not. The reason for this is simple: the vast majority of performance problems can be solved by implementing the proper algorithms, the ones with proven Big O-characteristics so you know the performance you'll get plus you know the algorithm will work. The time taken by the algorithm implementing code is inevitable: you already implemented the best algorithm. You might find some optimizations on the technical level but in general these are minor. Let's look at the four steps to see how they guide us through the quest to find and fix performance problems. Isolate The first thing you need to do is to isolate the areas in your application which are assumed to be slow. For example, if your application is a web application and a given page is taking several seconds or even minutes to load, it's a good candidate to check out. It's important to start with the isolate step because it allows you to focus on a single code path per area with a clear begin and end and ignore the rest. The rest of the steps are taken per identified problematic area. Keep in mind that isolation focuses on tasks in an application, not code snippets. A task is something that's started in your application by either another task or the user, or another program, and has a beginning and an end. You can see a task as a piece of functionality offered by your application.  Analyze Once you've determined the problem areas, you have to perform analysis on the code paths of each area, to see where the performance problems occur and which areas are not the problem. This is a multi-layered effort: an application which uses an O/R mapper typically consists of multiple parts: there's likely some kind of interface (web, webservice, windows etc.), a part which controls the interface and business logic, the O/R mapper part and the RDBMS, all connected with either a network or inter-process connections provided by the OS or other means. Each of these parts, including the connectivity plumbing, eat up a part of the total time it takes to complete a task, e.g. load a webpage with all orders of a given customer X. To understand which parts participate in the task / area we're investigating and how much they contribute to the total time taken to complete the task, analysis of each participating task is essential. Start with the code you wrote which starts the task, analyze the code and track the path it follows through your application. What does the code do along the way, verify whether it's correct or not. Analyze whether you have implemented the right algorithms in your code for this particular area. Remember we're looking at one area at a time, which means we're ignoring all other code paths, just the code path of the current problematic area, from begin to end and back. Don't dig in and start optimizing at the code level just yet. We're just analyzing. If your analysis reveals big architectural stupidity, it's perhaps a good idea to rethink the architecture at this point. For the rest, we're analyzing which means we collect data about what could be wrong, for each participating part of the complete application. Reviewing the code you wrote is a good tool to get deeper understanding of what is going on for a given task but ultimately it lacks precision and overview what really happens: humans aren't good code interpreters, computers are. We therefore need to utilize tools to get deeper understanding about which parts contribute how much time to the total task, triggered by which other parts and for example how many times are they called. There are two different kind of tools which are necessary: .NET profilers and O/R mapper / RDBMS profilers. .NET profiling .NET profilers (e.g. dotTrace by JetBrains or Ants by Red Gate software) show exactly which pieces of code are called, how many times they're called, and the time it took to run that piece of code, at the method level and sometimes even at the line level. The .NET profilers are essential tools for understanding whether the time taken to complete a given task / area in your application is consumed by .NET code, where exactly in your code, the path to that code, how many times that code was called by other code and thus reveals where hotspots are located: the areas where a solution can be found. Importantly, they also reveal which areas can be left alone: remember our penny wise pound foolish saying: if a profiler reveals that a group of methods are fast, or don't contribute much to the total time taken for a given task, ignore them. Even if the code in them is perhaps complex and looks like a candidate for optimization: you can work all day on that, it won't matter.  As we're focusing on a single area of the application, it's best to start profiling right before you actually activate the task/area. Most .NET profilers support this by starting the application without starting the profiling procedure just yet. You navigate to the particular part which is slow, start profiling in the profiler, in your application you perform the actions which are considered slow, and afterwards you get a snapshot in the profiler. The snapshot contains the data collected by the profiler during the slow action, so most data is produced by code in the area to investigate. This is important, because it allows you to stay focused on a single area. O/R mapper and RDBMS profiling .NET profilers give you a good insight in the .NET side of things, but not in the RDBMS side of the application. As this article is about O/R mapper powered applications, we're also looking at databases, and the software making it possible to consume the database in your application: the O/R mapper. To understand which parts of the O/R mapper and database participate how much to the total time taken for task T, we need different tools. There are two kind of tools focusing on O/R mappers and database performance profiling: O/R mapper profilers and RDBMS profilers. For O/R mapper profilers, you can look at LLBLGen Prof by hibernating rhinos or the Linq to Sql/LLBLGen Pro profiler by Huagati. Hibernating rhinos also have profilers for other O/R mappers like NHibernate (NHProf) and Entity Framework (EFProf) and work the same as LLBLGen Prof. For RDBMS profilers, you have to look whether the RDBMS vendor has a profiler. For example for SQL Server, the profiler is shipped with SQL Server, for Oracle it's build into the RDBMS, however there are also 3rd party tools. Which tool you're using isn't really important, what's important is that you get insight in which queries are executed during the task / area we're currently focused on and how long they took. Here, the O/R mapper profilers have an advantage as they collect the time it took to execute the query from the application's perspective so they also collect the time it took to transport data across the network. This is important because a query which returns a massive resultset or a resultset with large blob/clob/ntext/image fields takes more time to get transported across the network than a small resultset and a database profiler doesn't take this into account most of the time. Another tool to use in this case, which is more low level and not all O/R mappers support it (though LLBLGen Pro and NHibernate as well do) is tracing: most O/R mappers offer some form of tracing or logging system which you can use to collect the SQL generated and executed and often also other activity behind the scenes. While tracing can produce a tremendous amount of data in some cases, it also gives insight in what's going on. Interpret After we've completed the analysis step it's time to look at the data we've collected. We've done code reviews to see whether we've done anything stupid and which parts actually take place and if the proper algorithms have been implemented. We've done .NET profiling to see which parts are choke points and how much time they contribute to the total time taken to complete the task we're investigating. We've performed O/R mapper profiling and RDBMS profiling to see which queries were executed during the task, how many queries were generated and executed and how long they took to complete, including network transportation. All this data reveals two things: which parts are big contributors to the total time taken and which parts are irrelevant. Both aspects are very important. The parts which are irrelevant (i.e. don't contribute significantly to the total time taken) can be ignored from now on, we won't look at them. The parts which contribute a lot to the total time taken are important to look at. We now have to first look at the .NET profiler results, to see whether the time taken is consumed in our own code, in .NET framework code, in the O/R mapper itself or somewhere else. For example if most of the time is consumed by DbCommand.ExecuteReader, the time it took to complete the task is depending on the time the data is fetched from the database. If there was just 1 query executed, according to tracing or O/R mapper profilers / RDBMS profilers, check whether that query is optimal, uses indexes or has to deal with a lot of data. Interpret means that you follow the path from begin to end through the data collected and determine where, along the path, the most time is contributed. It also means that you have to check whether this was expected or is totally unexpected. My previous example of the 10 row resultset of a query which groups millions of rows will likely reveal that a long time is spend inside the database and almost no time is spend in the .NET code, meaning the RDBMS part contributes the most to the total time taken, the rest is compared to that time, irrelevant. Considering the vastness of the source data set, it's expected this will take some time. However, does it need tweaking? Perhaps all possible tweaks are already in place. In the interpret step you then have to decide that further action in this area is necessary or not, based on what the analysis results show: if the analysis results were unexpected and in the area where the most time is contributed to the total time taken is room for improvement, action should be taken. If not, you can only accept the situation and move on. In all cases, document your decision together with the analysis you've done. If you decide that the perceived performance problem is actually expected due to the nature of the task performed, it's essential that in the future when someone else looks at the application and starts asking questions you can answer them properly and new analysis is only necessary if situations changed. Fix After interpreting the analysis results you've concluded that some areas need adjustment. This is the fix step: you're actively correcting the performance problem with proper action targeted at the real cause. In many cases related to O/R mapper powered applications it means you'll use different features of the O/R mapper to achieve the same goal, or apply optimizations at the RDBMS level. It could also mean you apply caching inside your application (compromise memory consumption over performance) to avoid unnecessary re-querying data and re-consuming the results. After applying a change, it's key you re-do the analysis and interpretation steps: compare the results and expectations with what you had before, to see whether your actions had any effect or whether it moved the problem to a different part of the application. Don't fall into the trap to do partly analysis: do the full analysis again: .NET profiling and O/R mapper / RDBMS profiling. It might very well be that the changes you've made make one part faster but another part significantly slower, in such a way that the overall problem hasn't changed at all. Performance tuning is dealing with compromises and making choices: to use one feature over the other, to accept a higher memory footprint, to go away from the strict-OO path and execute queries directly onto the RDBMS, these are choices and compromises which will cross your path if you want to fix performance problems with respect to O/R mappers or data-access and databases in general. In most cases it's not a big issue: alternatives are often good choices too and the compromises aren't that hard to deal with. What is important is that you document why you made a choice, a compromise: which analysis data, which interpretation led you to the choice made. This is key for good maintainability in the years to come. Most common performance problems with O/R mappers Below is an incomplete list of common performance problems related to data-access / O/R mappers / RDBMS code. It will help you with fixing the hotspots you found in the interpretation step. SELECT N+1: (Lazy-loading specific). Lazy loading triggered performance bottlenecks. Consider a list of Orders bound to a grid. You have a Field mapped onto a related field in Order, Customer.CompanyName. Showing this column in the grid will make the grid fetch (indirectly) for each row the Customer row. This means you'll get for the single list not 1 query (for the orders) but 1+(the number of orders shown) queries. To solve this: use eager loading using a prefetch path to fetch the customers with the orders. SELECT N+1 is easy to spot with an O/R mapper profiler or RDBMS profiler: if you see a lot of identical queries executed at once, you have this problem. Prefetch paths using many path nodes or sorting, or limiting. Eager loading problem. Prefetch paths can help with performance, but as 1 query is fetched per node, it can be the number of data fetched in a child node is bigger than you think. Also consider that data in every node is merged on the client within the parent. This is fast, but it also can take some time if you fetch massive amounts of entities. If you keep fetches small, you can use tuning parameters like the ParameterizedPrefetchPathThreshold setting to get more optimal queries. Deep inheritance hierarchies of type Target Per Entity/Type. If you use inheritance of type Target per Entity / Type (each type in the inheritance hierarchy is mapped onto its own table/view), fetches will join subtype- and supertype tables in many cases, which can lead to a lot of performance problems if the hierarchy has many types. With this problem, keep inheritance to a minimum if possible, or switch to a hierarchy of type Target Per Hierarchy, which means all entities in the inheritance hierarchy are mapped onto the same table/view. Of course this has its own set of drawbacks, but it's a compromise you might want to take. Fetching massive amounts of data by fetching large lists of entities. LLBLGen Pro supports paging (and limiting the # of rows returned), which is often key to process through large sets of data. Use paging on the RDBMS if possible (so a query is executed which returns only the rows in the page requested). When using paging in a web application, be sure that you switch server-side paging on on the datasourcecontrol used. In this case, paging on the grid alone is not enough: this can lead to fetching a lot of data which is then loaded into the grid and paged there. Keep note that analyzing queries for paging could lead to the false assumption that paging doesn't occur, e.g. when the query contains a field of type ntext/image/clob/blob and DISTINCT can't be applied while it should have (e.g. due to a join): the datareader will do DISTINCT filtering on the client. this is a little slower but it does perform paging functionality on the data-reader so it won't fetch all rows even if the query suggests it does. Fetch massive amounts of data because blob/clob/ntext/image fields aren't excluded. LLBLGen Pro supports field exclusion for queries. You can exclude fields (also in prefetch paths) per query to avoid fetching all fields of an entity, e.g. when you don't need them for the logic consuming the resultset. Excluding fields can greatly reduce the amount of time spend on data-transport across the network. Use this optimization if you see that there's a big difference between query execution time on the RDBMS and the time reported by the .NET profiler for the ExecuteReader method call. Doing client-side aggregates/scalar calculations by consuming a lot of data. If possible, try to formulate a scalar query or group by query using the projection system or GetScalar functionality of LLBLGen Pro to do data consumption on the RDBMS server. It's far more efficient to process data on the RDBMS server than to first load it all in memory, then traverse the data in-memory to calculate a value. Using .ToList() constructs inside linq queries. It might be you use .ToList() somewhere in a Linq query which makes the query be run partially in-memory. Example: var q = from c in metaData.Customers.ToList() where c.Country=="Norway" select c; This will actually fetch all customers in-memory and do an in-memory filtering, as the linq query is defined on an IEnumerable<T>, and not on the IQueryable<T>. Linq is nice, but it can often be a bit unclear where some parts of a Linq query might run. Fetching all entities to delete into memory first. To delete a set of entities it's rather inefficient to first fetch them all into memory and then delete them one by one. It's more efficient to execute a DELETE FROM ... WHERE query on the database directly to delete the entities in one go. LLBLGen Pro supports this feature, and so do some other O/R mappers. It's not always possible to do this operation in the context of an O/R mapper however: if an O/R mapper relies on a cache, these kind of operations are likely not supported because they make it impossible to track whether an entity is actually removed from the DB and thus can be removed from the cache. Fetching all entities to update with an expression into memory first. Similar to the previous point: it is more efficient to update a set of entities directly with a single UPDATE query using an expression instead of fetching the entities into memory first and then updating the entities in a loop, and afterwards saving them. It might however be a compromise you don't want to take as it is working around the idea of having an object graph in memory which is manipulated and instead makes the code fully aware there's a RDBMS somewhere. Conclusion Performance tuning is almost always about compromises and making choices. It's also about knowing where to look and how the systems in play behave and should behave. The four steps I provided should help you stay focused on the real problem and lead you towards the solution. Knowing how to optimally use the systems participating in your own code (.NET framework, O/R mapper, RDBMS, network/services) is key for success as well as knowing what's going on inside the application you built. I hope you'll find this guide useful in tracking down performance problems and dealing with them in a useful way.  

    Read the article

  • Inside Red Gate - Ricky Leeks

    - by Simon Cooper
    So, one of our profilers has a problem. Red Gate produces two .NET profilers - ANTS Performance Profiler (APP) and ANTS Memory Profiler (AMP). Both products help .NET developers solve problems they are virtually guaranteed to encounter at some point in their careers - slow code, and high memory usage, respectively. Everyone understands slow code - the symptoms are very obvious (an operation takes 2 hours when it should take 10 seconds), you know when you've solved it (the same operation now takes 15 seconds), and everyone understands how you can use a profiler like APP to help solve your particular problem. High memory usage is a much more subtle and misunderstood concept. How can .NET have memory leaks? The garbage collector, and how the CLR uses and frees memory, is one of the most misunderstood concepts in .NET. There's hundreds of blog posts out there covering various aspects of the GC and .NET memory, some of them helpful, some of them confusing, and some of them are just plain wrong. There's a lot of misconceptions out there. And, if you have got an application that uses far too much memory, it can be hard to wade through all the contradictory information available to even get an idea as to what's going on, let alone trying to solve it. That's where a memory profiler, like AMP, comes into play. Unfortunately, that's not the end of the issue. .NET memory management is a large, complicated, and misunderstood problem. Even armed with a profiler, you need to understand what .NET is doing with your objects, how it processes them, and how it frees them, to be able to use the profiler effectively to solve your particular problem. And that's what's wrong with AMP - even with all the thought, designs, UX sessions, and research we've put into AMP itself, some users simply don't have the knowledge required to be able to understand what AMP is telling them about how their application uses memory, and so they have problems understanding & solving their memory problem. Ricky Leeks This is where Ricky Leeks comes in. Created by one of the many...colourful...people in Red Gate, he headlines and promotes several tutorials, pages, and articles all with information on how .NET memory management actually works, with the goal to help educate developers on .NET memory management. And educating us all on how far you can push various vegetable-based puns. This, in turn, not only helps them understand and solve any memory issues they may be having, but helps them proactively code against such memory issues in their existing code. Ricky's latest outing is an interview on .NET Rocks, providing information on the Top 5 .NET Memory Management Gotchas, along with information on a free ebook on .NET Memory Management. Don't worry, there's loads more vegetable-based jokes where those came from...

    Read the article

  • VS2010 crashes when opening a vsp generated using VS 2012

    - by Tarun Arora
    I recently profiled some web applications using Visual Studio 2012, a vsp (Visual Studio Profile) file was generated as a result of the profiling session. I could successfully open the vsp file in Visual Studio 2012 as expected but when I tried to open the vsp file in Visual Studio 2010 the VS2010 IDE crashed. As a responsible citizen I raised bug # 762202 on Microsoft Connect site using the Microsoft Visual Studio 2012 Feedback Client. Note – In case you didn’t already know, VSP generated in Visual Studio 2012 is not backward compatible. Please refer below for the steps to reproduce the issue and the resolution of the connect bug. 1. Behaviour and Steps to Reproduce the Issue Description I have generated a vsp file by using the Visual Studio 2012 Standalone profiler. When I try and open the vsp file in Visual Studio 2010 the IDE crashes. I understand that a vsp generated by using VS 2012 cannot be opened in VS 2010, but the IDE crashing is not the behaviour I would expect to see. Steps to Reproduce the Issue 1. Pick up the Stand lone profiler from the VS 2012 installation media. The folder has both x 64 and x86 installer, since the machine I am using is x64 bit. I have installed the x64 version of the standalone profiler. 2. I have configured the system path by setting the 'environment variable' path to where the profiler is installed. In my case this is, C:\Program Files (x86)\Microsoft Visual Studio 11.0\Team Tools\Performance Tools 3. Created a new environment variable _NT_SYMBOL_PATH and set its value to CACHE*C:\SYMBOLSCACHE;SRV*C:\SYMBOLSCACHE*HTTP://MSDL.MICROSOFT.COM/DOWNLOAD/SYMBOLS;\\FOO\BUILD1234 4. Open up CMD as an administrator and run 'VSPerfASPNETCmd /tip http://localhost:56180/ /o:C:\Temp\SampleEISK.vsp' 5. This generates the following message on the cmd       Microsoft (R) VSPerf ASP.NET Command, Version 11.0.0.0     Copyright (C) Microsoft Corporation. All rights reserved.     Configuring and attaching to ASP.NET process. Please wait.     Setting up profiling environment.     Starting monitor.     Launching ASP.NET service.     Attaching Monitor to process.     Launching Internet Explorer.     The profiler is attached to ASP.net. Please run your application scenario now.     Press Enter to stop data collection...   6. I perform certain actions and then I come back to the cmd and hit enter to shut down the profiling. Once I do this, the following message is written to the cmd, Press Enter to stop data collection... Profiling now shut down. Report file "C:\Temp\SampleEISK.vsp" was generated. Running VsPerfReport, packing symbols into the .VSP. Shutting down profiling and restarting ASP.NET. Please wait. Restarting w3wp.exe.   7. I look in the C:\Temp folder and I can see the SampleEISK.vsp file generated. I can successfully open this file in Visual Studio 2012. 8. When I am trying to open the vsp file in VS 2010 the VS 2010 IDE crashes. Kaboooom! What I would expect to happen I expect to receive a message "VS 2010 does not support the vsp file generated by VS 2012". What actually happened The VS 2010 IDE crashed 2. Resolution This is a valid bug! However, there isn’t much value in releasing a hotfix for this issue. Refer below to the resolution provided by the Visual Studio Profiler Team.  Thank you for taking the time to report this issue. We completely agree that Visual Studio 2010 should not crash. However in this particular case this is not a bug we are going to retroactively release a fix to 2010 for at this point. Given that a fix would not unblock the scenario of opening a 2012 created file on Visual Studio 2010, and there is not an active update channel for Visual Studio 2010 other than manually locating and installing hot fixes, we will not be fixing this particular issue. Best Regards, Visual Studio Profiler Team   Though it would be great to improve the behaviour however, this is not a defect that would stop you from progressing in any way. It’s important to note however that VSP files generated by Visual Studio 2012 are not backward compatible so you should refrain from opening these files in Visual Studio 2010.

    Read the article

  • Join our Marketing Intelligence Team in Dublin!

    - by jessica.ebbelaar
    Do you want to work with the brightest minds in the industry? Want to be part of a global team that’s changing the way the world does business? Then Oracle is the place for YOU. Join now as a Marketing Intelligence Representative. You will have the opportunity to develop within the role through working alongside the Business Development, Sales and Marketing teams within Oracle. The Marketing Intelligence Group is viewed as a true talent pool for the Business Development and Sales Teams. Oracle offers a structured training programme for Marketing Intelligence Representatives and Business Development Consultants including our approved sales certified training methodology along with regular product training. Miriam started her career as a Marketing Intelligence Representative six years ago, and shares what she has learned and how her career is progressing. My Career Path at Oracle: June 2005 – October 2005: Profiler in the Marketing Intelligence Team November 2005 - October 2006: Team Leader for MIT November 2006 - February 2008: Business Development Consultant Iberia March 2008 - December 2010: Lead Management Specialist Currently: Sales Program Manager for Iberia & Benelux What did you learn from your role in the Market Intelligence Team Being a Profiler helped me to understand how an organisation works, from the beginning to the end. It is like being in University but being paid! The three key things I learnt in this role are: Knowledge of customers: You are on the phone with over 70 customers daily. Not only does this give you an overview of the IT infrastructure of the customers companies but also how to manage their questions and rejections. Essentially you are learning how to convert their pain and complaints into business opportunities. Knowledge of Oracle: As a Profiler you get an excellent overview of how Oracle works internally, from Marketing to Sales, without forgetting the Operations Team. Knowledge about yourself: As a Profiler I learnt how to work outside of my comfort zone, there is a new challenge almost every day but Oracle are there to support you every step of the way. Oracle really invests in developing the MIT Team and as a Profiler you can expect product and sales training on a monthly basis. How did you progress from MIT to Business Development Group (BDG)? I made sure that my manager knew from the very beginning that I was keen to progress at Oracle and I was set very clear objectives to help me reach my goal.  My manager was very supportive and ensured I received all the training I needed. After I became a Team Leader of Profiling, I moved to an Iberia BDG position. How you feel your experience in MI has helped you in your current role? I truly believe that the MI position gives you a great overview of Oracle and this has really helped me in my current position.  I am the Sales Program Manager for IBERIA & Benelux and in my campaigns I need to target the right companies and the right job specs.  My time in the Market Intelligence team really helped me to understand how to focus and target my campaigns so I know I don’t miss any business opportunities! How would you sum up your Oracle experience? Oracle is a big organisation with big opportunities. With the right skills and with the great training programs that Oracle offer, the only limit is you! If you have any questions related to this article feel free to contact [email protected] You can find all our job opportunities via http://campus.oracle.com. Tags van Technorati: Marketing Intelligence,Benelux,Iberia,Profiler,Business Development,Sales Representatives,BDG,Business Development Group,opportunities,Oracle

    Read the article

  • Analysing and measuring the performance of a .NET application (survey results)

    - by Laila
    Back in December last year, I asked myself: could it be that .NET developers think that you need three days and a PhD to do performance profiling on their code? What if developers are shunning profilers because they perceive them as too complex to use? If so, then what method do they use to measure and analyse the performance of their .NET applications? Do they even care about performance? So, a few weeks ago, I decided to get a 1-minute survey up and running in the hopes that some good, hard data would clear the matter up once and for all. I posted the survey on Simple Talk and got help from a few people to promote it. The survey consisted of 3 simple questions: Amazingly, 533 developers took the time to respond - which means I had enough data to get representative results! So before I go any further, I would like to thank all of you who contributed, because I now have some pretty good answers to the troubling questions I was asking myself. To thank you properly, I thought I would share some of the results with you. First of all, application performance is indeed important to most of you. In fact, performance is an intrinsic part of the development cycle for a good 40% of you, which is much higher than I had anticipated, I have to admit. (I know, "Have a little faith Laila!") When asked what tool you use to measure and analyse application performance, I found that nearly half of the respondents use logging statements, a third use performance counters, and 70% of respondents use a profiler of some sort (a 3rd party performance profilers, the CLR profiler or the Visual Studio profiler). The importance attributed to logging statements did surprise me a little. I am still not sure why somebody would go to the trouble of manually instrumenting code in order to measure its performance, instead of just using a profiler. I personally find the process of annotating code, calculating times from log files, and relating it all back to your source terrifyingly laborious. Not to mention that you then need to remember to turn it all off later! Even when you have logging in place throughout all your code anyway, you still have a fair amount of potentially error-prone calculation to sift through the results; in addition, you'll only get method-level rather than line-level timings, and you won't get timings from any framework or library methods you don't have source for. To top it all, we all know that bottlenecks are rarely where you would expect them to be, so you could be wasting time looking for a performance problem in the wrong place. On the other hand, profilers do all the work for you: they automatically collect the CPU and wall-clock timings, and present the results from method timing all the way down to individual lines of code. Maybe I'm missing a trick. I would love to know about the types of scenarios where you actively prefer to use logging statements. Finally, while a third of the respondents didn't have a strong opinion about code performance profilers, those who had an opinion thought that they were mainly complex to use and time consuming. Three respondents in particular summarised this perfectly: "sometimes, they are rather complex to use, adding an additional time-sink to the process of trying to resolve the existing problem". "they are simple to use, but the results are hard to understand" "Complex to find the more advanced things, easy to find some low hanging fruit". These results confirmed my suspicions: Profilers are seen to be designed for more advanced users who can use them effectively and make sense of the results. I found yet more interesting information when I started comparing samples of "developers for whom performance is an important part of the dev cycle", with those "to whom performance is only looked at in times of crisis", and "developers to whom performance is not important, as long as the app works". See the three graphs below. Sample of developers to whom performance is an important part of the dev cycle: Sample of developers to whom performance is important only in times of crisis: Sample of developers to whom performance is not important, as long as the app works: As you can see, there is a strong correlation between the usage of a profiler and the importance attributed to performance: indeed, the more important performance is to a development team, the more likely they are to use a profiler. In addition, developers to whom performance is an important part of the dev cycle have a higher tendency to use a much wider range of methods for performance measurement and analysis. And, unsurprisingly, the less important performance is, the less varied the methods of measurement are. So all in all, to come back to my random questions: .NET developers do care about performance. Those who care the most use a wider range of performance measurement methods than those who care less. But overall, logging statements, performance counters and third party performance profilers are the performance measurement methods of choice for most developers. Finally, although most of you find code profilers complex to use, those of you who care the most about performance tend to use profilers more than those of you to whom performance is not so important.

    Read the article

  • Will JVisualVM degrade application performance?

    - by rocky
    I have doubts in JVisual VM profiler tool related to performance. I have requirement to implement a JVM Monitoring tool for my enterpise java application. I have gone through some profiling tools in market but all them are having some kind of agent file which we need include in server startup. I have a fear that these client agent will degrade my application performance will more. So I have decided to JVisual VM because this profiler tool comes with JDK itself but before implementing JVisualVM, does anybody faces any issues with JVisualVM profiler tool? As well as, is this safe if I implement in application?

    Read the article

  • More CPU cores may not always lead to better performance – MAXDOP and query memory distribution in spotlight

    - by sqlworkshops
    More hardware normally delivers better performance, but there are exceptions where it can hinder performance. Understanding these exceptions and working around it is a major part of SQL Server performance tuning.   When a memory allocating query executes in parallel, SQL Server distributes memory to each task that is executing part of the query in parallel. In our example the sort operator that executes in parallel divides the memory across all tasks assuming even distribution of rows. Common memory allocating queries are that perform Sort and do Hash Match operations like Hash Join or Hash Aggregation or Hash Union.   In reality, how often are column values evenly distributed, think about an example; are employees working for your company distributed evenly across all the Zip codes or mainly concentrated in the headquarters? What happens when you sort result set based on Zip codes? Do all products in the catalog sell equally or are few products hot selling items?   One of my customers tested the below example on a 24 core server with various MAXDOP settings and here are the results:MAXDOP 1: CPU time = 1185 ms, elapsed time = 1188 msMAXDOP 4: CPU time = 1981 ms, elapsed time = 1568 msMAXDOP 8: CPU time = 1918 ms, elapsed time = 1619 msMAXDOP 12: CPU time = 2367 ms, elapsed time = 2258 msMAXDOP 16: CPU time = 2540 ms, elapsed time = 2579 msMAXDOP 20: CPU time = 2470 ms, elapsed time = 2534 msMAXDOP 0: CPU time = 2809 ms, elapsed time = 2721 ms - all 24 cores.In the above test, when the data was evenly distributed, the elapsed time of parallel query was always lower than serial query.   Why does the query get slower and slower with more CPU cores / higher MAXDOP? Maybe you can answer this question after reading the article; let me know: [email protected].   Well you get the point, let’s see an example.   The best way to learn is to practice. To create the below tables and reproduce the behavior, join the mailing list by using this link: www.sqlworkshops.com/ml and I will send you the table creation script.   Let’s update the Employees table with 49 out of 50 employees located in Zip code 2001. update Employees set Zip = EmployeeID / 400 + 1 where EmployeeID % 50 = 1 update Employees set Zip = 2001 where EmployeeID % 50 != 1 go update statistics Employees with fullscan go   Let’s create the temporary table #FireDrill with all possible Zip codes. drop table #FireDrill go create table #FireDrill (Zip int primary key) insert into #FireDrill select distinct Zip from Employees update statistics #FireDrill with fullscan go  Let’s execute the query serially with MAXDOP 1. --Example provided by www.sqlworkshops.com --Execute query with uneven Zip code distribution --First serially with MAXDOP 1 set statistics time on go declare @EmployeeID int, @EmployeeName varchar(48),@zip int select @EmployeeName = e.EmployeeName, @zip = e.Zip from Employees e       inner join #FireDrill fd on (e.Zip = fd.Zip)       order by e.Zip option (maxdop 1) goThe query took 1011 ms to complete.   The execution plan shows the 77816 KB of memory was granted while the estimated rows were 799624.  No Sort Warnings in SQL Server Profiler.  Now let’s execute the query in parallel with MAXDOP 0. --Example provided by www.sqlworkshops.com --Execute query with uneven Zip code distribution --In parallel with MAXDOP 0 set statistics time on go declare @EmployeeID int, @EmployeeName varchar(48),@zip int select @EmployeeName = e.EmployeeName, @zip = e.Zip from Employees e       inner join #FireDrill fd on (e.Zip = fd.Zip)       order by e.Zip option (maxdop 0) go The query took 1912 ms to complete.  The execution plan shows the 79360 KB of memory was granted while the estimated rows were 799624.  The estimated number of rows between serial and parallel plan are the same. The parallel plan has slightly more memory granted due to additional overhead. Sort properties shows the rows are unevenly distributed over the 4 threads.   Sort Warnings in SQL Server Profiler.   Intermediate Summary: The reason for the higher duration with parallel plan was sort spill. This is due to uneven distribution of employees over Zip codes, especially concentration of 49 out of 50 employees in Zip code 2001. Now let’s update the Employees table and distribute employees evenly across all Zip codes.   update Employees set Zip = EmployeeID / 400 + 1 go update statistics Employees with fullscan go  Let’s execute the query serially with MAXDOP 1. --Example provided by www.sqlworkshops.com --Execute query with uneven Zip code distribution --Serially with MAXDOP 1 set statistics time on go declare @EmployeeID int, @EmployeeName varchar(48),@zip int select @EmployeeName = e.EmployeeName, @zip = e.Zip from Employees e       inner join #FireDrill fd on (e.Zip = fd.Zip)       order by e.Zip option (maxdop 1) go   The query took 751 ms to complete.  The execution plan shows the 77816 KB of memory was granted while the estimated rows were 784707.  No Sort Warnings in SQL Server Profiler.   Now let’s execute the query in parallel with MAXDOP 0. --Example provided by www.sqlworkshops.com --Execute query with uneven Zip code distribution --In parallel with MAXDOP 0 set statistics time on go declare @EmployeeID int, @EmployeeName varchar(48),@zip int select @EmployeeName = e.EmployeeName, @zip = e.Zip from Employees e       inner join #FireDrill fd on (e.Zip = fd.Zip)       order by e.Zip option (maxdop 0) go The query took 661 ms to complete.  The execution plan shows the 79360 KB of memory was granted while the estimated rows were 784707.  Sort properties shows the rows are evenly distributed over the 4 threads. No Sort Warnings in SQL Server Profiler.    Intermediate Summary: When employees were distributed unevenly, concentrated on 1 Zip code, parallel sort spilled while serial sort performed well without spilling to tempdb. When the employees were distributed evenly across all Zip codes, parallel sort and serial sort did not spill to tempdb. This shows uneven data distribution may affect the performance of some parallel queries negatively. For detailed discussion of memory allocation, refer to webcasts available at www.sqlworkshops.com/webcasts.     Some of you might conclude from the above execution times that parallel query is not faster even when there is no spill. Below you can see when we are joining limited amount of Zip codes, parallel query will be fasted since it can use Bitmap Filtering.   Let’s update the Employees table with 49 out of 50 employees located in Zip code 2001. update Employees set Zip = EmployeeID / 400 + 1 where EmployeeID % 50 = 1 update Employees set Zip = 2001 where EmployeeID % 50 != 1 go update statistics Employees with fullscan go  Let’s create the temporary table #FireDrill with limited Zip codes. drop table #FireDrill go create table #FireDrill (Zip int primary key) insert into #FireDrill select distinct Zip       from Employees where Zip between 1800 and 2001 update statistics #FireDrill with fullscan go  Let’s execute the query serially with MAXDOP 1. --Example provided by www.sqlworkshops.com --Execute query with uneven Zip code distribution --Serially with MAXDOP 1 set statistics time on go declare @EmployeeID int, @EmployeeName varchar(48),@zip int select @EmployeeName = e.EmployeeName, @zip = e.Zip from Employees e       inner join #FireDrill fd on (e.Zip = fd.Zip)       order by e.Zip option (maxdop 1) go The query took 989 ms to complete.  The execution plan shows the 77816 KB of memory was granted while the estimated rows were 785594. No Sort Warnings in SQL Server Profiler.  Now let’s execute the query in parallel with MAXDOP 0. --Example provided by www.sqlworkshops.com --Execute query with uneven Zip code distribution --In parallel with MAXDOP 0 set statistics time on go declare @EmployeeID int, @EmployeeName varchar(48),@zip int select @EmployeeName = e.EmployeeName, @zip = e.Zip from Employees e       inner join #FireDrill fd on (e.Zip = fd.Zip)       order by e.Zip option (maxdop 0) go The query took 1799 ms to complete.  The execution plan shows the 79360 KB of memory was granted while the estimated rows were 785594.  Sort Warnings in SQL Server Profiler.    The estimated number of rows between serial and parallel plan are the same. The parallel plan has slightly more memory granted due to additional overhead.  Intermediate Summary: The reason for the higher duration with parallel plan even with limited amount of Zip codes was sort spill. This is due to uneven distribution of employees over Zip codes, especially concentration of 49 out of 50 employees in Zip code 2001.   Now let’s update the Employees table and distribute employees evenly across all Zip codes. update Employees set Zip = EmployeeID / 400 + 1 go update statistics Employees with fullscan go Let’s execute the query serially with MAXDOP 1. --Example provided by www.sqlworkshops.com --Execute query with uneven Zip code distribution --Serially with MAXDOP 1 set statistics time on go declare @EmployeeID int, @EmployeeName varchar(48),@zip int select @EmployeeName = e.EmployeeName, @zip = e.Zip from Employees e       inner join #FireDrill fd on (e.Zip = fd.Zip)       order by e.Zip option (maxdop 1) go The query took 250  ms to complete.  The execution plan shows the 9016 KB of memory was granted while the estimated rows were 79973.8.  No Sort Warnings in SQL Server Profiler.  Now let’s execute the query in parallel with MAXDOP 0.  --Example provided by www.sqlworkshops.com --Execute query with uneven Zip code distribution --In parallel with MAXDOP 0 set statistics time on go declare @EmployeeID int, @EmployeeName varchar(48),@zip int select @EmployeeName = e.EmployeeName, @zip = e.Zip from Employees e       inner join #FireDrill fd on (e.Zip = fd.Zip)       order by e.Zip option (maxdop 0) go The query took 85 ms to complete.  The execution plan shows the 13152 KB of memory was granted while the estimated rows were 784707.  No Sort Warnings in SQL Server Profiler.    Here you see, parallel query is much faster than serial query since SQL Server is using Bitmap Filtering to eliminate rows before the hash join.   Parallel queries are very good for performance, but in some cases it can hinder performance. If one identifies the reason for these hindrances, then it is possible to get the best out of parallelism. I covered many aspects of monitoring and tuning parallel queries in webcasts (www.sqlworkshops.com/webcasts) and articles (www.sqlworkshops.com/articles). I suggest you to watch the webcasts and read the articles to better understand how to identify and tune parallel query performance issues.   Summary: One has to avoid sort spill over tempdb and the chances of spills are higher when a query executes in parallel with uneven data distribution. Parallel query brings its own advantage, reduced elapsed time and reduced work with Bitmap Filtering. So it is important to understand how to avoid spills over tempdb and when to execute a query in parallel.   I explain these concepts with detailed examples in my webcasts (www.sqlworkshops.com/webcasts), I recommend you to watch them. The best way to learn is to practice. To create the above tables and reproduce the behavior, join the mailing list at www.sqlworkshops.com/ml and I will send you the relevant SQL Scripts.   Register for the upcoming 3 Day Level 400 Microsoft SQL Server 2008 and SQL Server 2005 Performance Monitoring & Tuning Hands-on Workshop in London, United Kingdom during March 15-17, 2011, click here to register / Microsoft UK TechNet.These are hands-on workshops with a maximum of 12 participants and not lectures. For consulting engagements click here.   Disclaimer and copyright information:This article refers to organizations and products that may be the trademarks or registered trademarks of their various owners. Copyright of this article belongs to R Meyyappan / www.sqlworkshops.com. You may freely use the ideas and concepts discussed in this article with acknowledgement (www.sqlworkshops.com), but you may not claim any of it as your own work. This article is for informational purposes only; you use any of the suggestions given here entirely at your own risk.   Register for the upcoming 3 Day Level 400 Microsoft SQL Server 2008 and SQL Server 2005 Performance Monitoring & Tuning Hands-on Workshop in London, United Kingdom during March 15-17, 2011, click here to register / Microsoft UK TechNet.These are hands-on workshops with a maximum of 12 participants and not lectures. For consulting engagements click here.   R Meyyappan [email protected] LinkedIn: http://at.linkedin.com/in/rmeyyappan  

    Read the article

  • Plan Caching and Query Memory Part II (Hash Match) – When not to use stored procedure - Most common performance mistake SQL Server developers make.

    - by sqlworkshops
    SQL Server estimates Memory requirement at compile time, when stored procedure or other plan caching mechanisms like sp_executesql or prepared statement are used, the memory requirement is estimated based on first set of execution parameters. This is a common reason for spill over tempdb and hence poor performance. Common memory allocating queries are that perform Sort and do Hash Match operations like Hash Join or Hash Aggregation or Hash Union. This article covers Hash Match operations with examples. It is recommended to read Plan Caching and Query Memory Part I before this article which covers an introduction and Query memory for Sort. In most cases it is cheaper to pay for the compilation cost of dynamic queries than huge cost for spill over tempdb, unless memory requirement for a query does not change significantly based on predicates.   This article covers underestimation / overestimation of memory for Hash Match operation. Plan Caching and Query Memory Part I covers underestimation / overestimation for Sort. It is important to note that underestimation of memory for Sort and Hash Match operations lead to spill over tempdb and hence negatively impact performance. Overestimation of memory affects the memory needs of other concurrently executing queries. In addition, it is important to note, with Hash Match operations, overestimation of memory can actually lead to poor performance.   To read additional articles I wrote click here.   The best way to learn is to practice. To create the below tables and reproduce the behavior, join the mailing list by using this link: www.sqlworkshops.com/ml and I will send you the table creation script. Most of these concepts are also covered in our webcasts: www.sqlworkshops.com/webcasts  Let’s create a Customer’s State table that has 99% of customers in NY and the rest 1% in WA.Customers table used in Part I of this article is also used here.To observe Hash Warning, enable 'Hash Warning' in SQL Profiler under Events 'Errors and Warnings'. --Example provided by www.sqlworkshops.com drop table CustomersState go create table CustomersState (CustomerID int primary key, Address char(200), State char(2)) go insert into CustomersState (CustomerID, Address) select CustomerID, 'Address' from Customers update CustomersState set State = 'NY' where CustomerID % 100 != 1 update CustomersState set State = 'WA' where CustomerID % 100 = 1 go update statistics CustomersState with fullscan go   Let’s create a stored procedure that joins customers with CustomersState table with a predicate on State. --Example provided by www.sqlworkshops.com create proc CustomersByState @State char(2) as begin declare @CustomerID int select @CustomerID = e.CustomerID from Customers e inner join CustomersState es on (e.CustomerID = es.CustomerID) where es.State = @State option (maxdop 1) end go  Let’s execute the stored procedure first with parameter value ‘WA’ – which will select 1% of data. set statistics time on go --Example provided by www.sqlworkshops.com exec CustomersByState 'WA' goThe stored procedure took 294 ms to complete.  The stored procedure was granted 6704 KB based on 8000 rows being estimated.  The estimated number of rows, 8000 is similar to actual number of rows 8000 and hence the memory estimation should be ok.  There was no Hash Warning in SQL Profiler. To observe Hash Warning, enable 'Hash Warning' in SQL Profiler under Events 'Errors and Warnings'.   Now let’s execute the stored procedure with parameter value ‘NY’ – which will select 99% of data. -Example provided by www.sqlworkshops.com exec CustomersByState 'NY' go  The stored procedure took 2922 ms to complete.   The stored procedure was granted 6704 KB based on 8000 rows being estimated.    The estimated number of rows, 8000 is way different from the actual number of rows 792000 because the estimation is based on the first set of parameter value supplied to the stored procedure which is ‘WA’ in our case. This underestimation will lead to spill over tempdb, resulting in poor performance.   There was Hash Warning (Recursion) in SQL Profiler. To observe Hash Warning, enable 'Hash Warning' in SQL Profiler under Events 'Errors and Warnings'.   Let’s recompile the stored procedure and then let’s first execute the stored procedure with parameter value ‘NY’.  In a production instance it is not advisable to use sp_recompile instead one should use DBCC FREEPROCCACHE (plan_handle). This is due to locking issues involved with sp_recompile, refer to our webcasts, www.sqlworkshops.com/webcasts for further details.   exec sp_recompile CustomersByState go --Example provided by www.sqlworkshops.com exec CustomersByState 'NY' go  Now the stored procedure took only 1046 ms instead of 2922 ms.   The stored procedure was granted 146752 KB of memory. The estimated number of rows, 792000 is similar to actual number of rows of 792000. Better performance of this stored procedure execution is due to better estimation of memory and avoiding spill over tempdb.   There was no Hash Warning in SQL Profiler.   Now let’s execute the stored procedure with parameter value ‘WA’. --Example provided by www.sqlworkshops.com exec CustomersByState 'WA' go  The stored procedure took 351 ms to complete, higher than the previous execution time of 294 ms.    This stored procedure was granted more memory (146752 KB) than necessary (6704 KB) based on parameter value ‘NY’ for estimation (792000 rows) instead of parameter value ‘WA’ for estimation (8000 rows). This is because the estimation is based on the first set of parameter value supplied to the stored procedure which is ‘NY’ in this case. This overestimation leads to poor performance of this Hash Match operation, it might also affect the performance of other concurrently executing queries requiring memory and hence overestimation is not recommended.     The estimated number of rows, 792000 is much more than the actual number of rows of 8000.  Intermediate Summary: This issue can be avoided by not caching the plan for memory allocating queries. Other possibility is to use recompile hint or optimize for hint to allocate memory for predefined data range.Let’s recreate the stored procedure with recompile hint. --Example provided by www.sqlworkshops.com drop proc CustomersByState go create proc CustomersByState @State char(2) as begin declare @CustomerID int select @CustomerID = e.CustomerID from Customers e inner join CustomersState es on (e.CustomerID = es.CustomerID) where es.State = @State option (maxdop 1, recompile) end go  Let’s execute the stored procedure initially with parameter value ‘WA’ and then with parameter value ‘NY’. --Example provided by www.sqlworkshops.com exec CustomersByState 'WA' go exec CustomersByState 'NY' go  The stored procedure took 297 ms and 1102 ms in line with previous optimal execution times.   The stored procedure with parameter value ‘WA’ has good estimation like before.   Estimated number of rows of 8000 is similar to actual number of rows of 8000.   The stored procedure with parameter value ‘NY’ also has good estimation and memory grant like before because the stored procedure was recompiled with current set of parameter values.  Estimated number of rows of 792000 is similar to actual number of rows of 792000.    The compilation time and compilation CPU of 1 ms is not expensive in this case compared to the performance benefit.   There was no Hash Warning in SQL Profiler.   Let’s recreate the stored procedure with optimize for hint of ‘NY’. --Example provided by www.sqlworkshops.com drop proc CustomersByState go create proc CustomersByState @State char(2) as begin declare @CustomerID int select @CustomerID = e.CustomerID from Customers e inner join CustomersState es on (e.CustomerID = es.CustomerID) where es.State = @State option (maxdop 1, optimize for (@State = 'NY')) end go  Let’s execute the stored procedure initially with parameter value ‘WA’ and then with parameter value ‘NY’. --Example provided by www.sqlworkshops.com exec CustomersByState 'WA' go exec CustomersByState 'NY' go  The stored procedure took 353 ms with parameter value ‘WA’, this is much slower than the optimal execution time of 294 ms we observed previously. This is because of overestimation of memory. The stored procedure with parameter value ‘NY’ has optimal execution time like before.   The stored procedure with parameter value ‘WA’ has overestimation of rows because of optimize for hint value of ‘NY’.   Unlike before, more memory was estimated to this stored procedure based on optimize for hint value ‘NY’.    The stored procedure with parameter value ‘NY’ has good estimation because of optimize for hint value of ‘NY’. Estimated number of rows of 792000 is similar to actual number of rows of 792000.   Optimal amount memory was estimated to this stored procedure based on optimize for hint value ‘NY’.   There was no Hash Warning in SQL Profiler.   This article covers underestimation / overestimation of memory for Hash Match operation. Plan Caching and Query Memory Part I covers underestimation / overestimation for Sort. It is important to note that underestimation of memory for Sort and Hash Match operations lead to spill over tempdb and hence negatively impact performance. Overestimation of memory affects the memory needs of other concurrently executing queries. In addition, it is important to note, with Hash Match operations, overestimation of memory can actually lead to poor performance.   Summary: Cached plan might lead to underestimation or overestimation of memory because the memory is estimated based on first set of execution parameters. It is recommended not to cache the plan if the amount of memory required to execute the stored procedure has a wide range of possibilities. One can mitigate this by using recompile hint, but that will lead to compilation overhead. However, in most cases it might be ok to pay for compilation rather than spilling sort over tempdb which could be very expensive compared to compilation cost. The other possibility is to use optimize for hint, but in case one sorts more data than hinted by optimize for hint, this will still lead to spill. On the other side there is also the possibility of overestimation leading to unnecessary memory issues for other concurrently executing queries. In case of Hash Match operations, this overestimation of memory might lead to poor performance. When the values used in optimize for hint are archived from the database, the estimation will be wrong leading to worst performance, so one has to exercise caution before using optimize for hint, recompile hint is better in this case.   I explain these concepts with detailed examples in my webcasts (www.sqlworkshops.com/webcasts), I recommend you to watch them. The best way to learn is to practice. To create the above tables and reproduce the behavior, join the mailing list at www.sqlworkshops.com/ml and I will send you the relevant SQL Scripts.  Register for the upcoming 3 Day Level 400 Microsoft SQL Server 2008 and SQL Server 2005 Performance Monitoring & Tuning Hands-on Workshop in London, United Kingdom during March 15-17, 2011, click here to register / Microsoft UK TechNet.These are hands-on workshops with a maximum of 12 participants and not lectures. For consulting engagements click here.   Disclaimer and copyright information:This article refers to organizations and products that may be the trademarks or registered trademarks of their various owners. Copyright of this article belongs to R Meyyappan / www.sqlworkshops.com. You may freely use the ideas and concepts discussed in this article with acknowledgement (www.sqlworkshops.com), but you may not claim any of it as your own work. This article is for informational purposes only; you use any of the suggestions given here entirely at your own risk.   R Meyyappan [email protected] LinkedIn: http://at.linkedin.com/in/rmeyyappan

    Read the article

< Previous Page | 3 4 5 6 7 8 9 10 11 12 13 14  | Next Page >