Intel® VTune™ Amplifier XE 2013

Performance profiler for serial and parallel performance analysis.

  • Collect a rich set of data to tune CPU & GPU compute performance, multi-core scalability, bandwidth and more
  • Sort, filter and visualize results for quick insight into performance bottlenecks
  • Automate regression tests and collect data remotely using the powerful command line

From $899
Buy Now

Or Download a Free 30-Day Evaluation Version

Service Pack 1 Released - What’s New

Optimize Serial and Parallel Performance

Intel® VTune™ Amplifier XE 2013 is the premier performance profiler for C, C++, C#, Fortran, Assembly and Java*.


Functions using the most CPU time float to the top of the list. Click on a function to see the call stack. Double click to see the source with detailed profile data.

Easy
Performance optimization can be difficult, but the performance profiling tool you use shouldn’t be.

Versatile – Rich Set of Performance Profiles
Whether you are tuning for the first time or doing advanced performance optimization, VTune Amplifier XE 2013 provides the data needed to meet a wide variety of tuning needs.  Collect a rich set of performance data for hotspots, threading, locks & waits, DirectX*, bandwidth and more.

Productive – Sort, Filter and Visualize
Good data is not enough.  You need tools to mine the data and make it easy to understand.  Powerful analysis lets you sort, filter and visualize results on the timeline and on your source.

New for 2013! 
Caller/callee, OpenCL, OpenMP 4.0, hardware stack sampling, better bandwidth analysis, Java profiling, tune Intel® Xeon Phi™ products, user tasks, DirectX* frames, call counts and more.


Quotes

"We achieved a significant improvement (almost 2x) even on one core by optimizing the code based on the information provided by Intel® VTune™ Amplifier XE. Good scalability is a result of usage of combination of Intel® TBB and OpenMP parallelization techniques. We achieved over 8x the performance of the previous version on 8 cores and almost 11x the performance on 16 cores."
Alexey Andrianov, R&D Director Deputy, Mechanical Analysis Division, Mentor Graphics Corporation

"Intel® VTune™ Amplifier XE analyzes complex code and helps us identify bottlenecks rapidly. By using it and other Intel® Software Development Tools, we were able to improve PIPESIM performance up to 10 times compared with the previous software version."
Rodney Lessard, Senior Scientist, Schlumberger

“The new VTune™ Amplifier XE brings even more capability to an already indispensable tool. The sampling based call stack hotspots is excellent and alone is worthy of the upgrade. We have also been impressed by how the concurrency and Locks and Waits analysis can even provide useful data on complex applications such as Premiere Pro.”
Rich Gerber - Engineering Manager, MediaCore, Adobe Systems Inc.

“The new interface is a joy to use. Intel® VTune Amplifier XE gives us precise, down-to-the-metal performance data that’s invaluable for pinpointing hotspots and evaluating the effect of optimizations”
Daniel Schwarz, Performance Engineer, Nik Software

“Intel® VTune™ Amplifier XE’s timeline is very information intensive.  It organizes the data I need to tune threaded applications.”
Sergey Zaritchny, Software Development Manager, Open Cascade SAS

“Last week, Intel® VTune™ Amplifier XE helped us find almost 3X performance improvement.  This week it helped us improve the performance another 3X.”
Claire Cates, Principal Developer, SAS Institute Inc.

“One of Intel® VTune™ Amplifier XE’s best features is that it is easy to use.  I did not need to read the documentation.”
Richard Shepherd, Software Engineer, ESRI (UK) Limited

Quickly Locate Code Taking A Lot of CPU Time (or GPU time)

Hotspots analysis gives you a sorted list of the functions using a lot of CPU time. This is where tuning will give you the biggest benefit. Click [+] for the call stacks. Double click to see the source.

New! On newer processors, optionally collect GPU data for tuning OpenCL applications. Correlate GPU and CPU activities. (Windows* only.)

See the Results on Your Source

A double click from the function list takes you to the hottest spot in the function.

Tune Threading with Locks and Waits Analysis

Quickly find a common cause of slow performance in parallel programs: waiting too long on a lock while the cores are underutilized during the wait. Profiles like "basic hotspots" and "locks & waits" use a software collector that works on both Intel and compatible processors. New! OpenMP 4.0 support.

Mine the Data with Timeline Filtering

Select a time range in the timeline to filter out data (e.g., application startup) that masks the information you need. When you select and filter in the timeline, the grid that lists functions using a lot of CPU time updates to show the list filtered for the selected time.

Visualize Thread Behavior

See when threads are running and waiting, and when transitions occur. Balance workloads. Find lock contention.

New!Profile Remote Systems

Configure your host system to collect data from a remote Linux target.

Low Overhead / High Resolution Hardware Profiling

In addition to "basic hotspots" analysis that works on both Intel and compatible processors, VTune Amplifier XE 2013 has "advanced hotspots" analysis that uses the Performance Monitoring Unit (PMU) on Intel processors to collect data with very low overhead. Increased resolution (~1 ms vs. ~10 ms) can find hot spots in small functions that run quickly. New! Now with optional stack collection to identify the calling sequence.

Advanced Analysis Like Bandwidth

Preset profiles provide an easy "point and shoot" set-up. Choose Hotspot, Lightweight Hotspot, Concurrency, Locks and Waits or more advanced analyses. No memorizing complex event names. Advanced profiles like memory bandwidth analysis, memory access and branch mispredictions find tuning opportunities. New! Advanced profiles can optionally collect stacks to identify the calling sequence. (Profiles vary by microarchitecture.)

Opportunities Highlighted

The cell is highlighted in pink when there is a potential tuning opportunity. Hover to get suggestions.

New! OpenMP Scalability Analysis

Visualize time regions from the fork point to the join point for each parallel region. See what is serial, what is balanced and what is imbalanced. Here we see 13.671 seconds in an imbalanced region, 3.652 seconds in a fairly well balanced region.

New! Tune OpenCL

On newer processors, optionally collect GPU data for tuning OpenCL applications. Correlate GPU and CPU activities. (Windows* only.)

No special builds

Use a production build with symbols from your normal compiler.

Low overhead

Accurate results you can count on.

Command line

Automate regression analysis. Simple remote collection.

System Wide Analysis

Tune drivers, kernel modules and multi-process apps.

New!Tune Inlining with Call Counts

When a function is called frequently it may make sense to "inline" the code and eliminate the overhead of the function call. VTune Amplifier XE 2013 now provides statistical call count data to help you make better inlining decisions. It also displays profile results on the source code, even if the code is inlined, making it easier to interpret profile results.

New!Auto Detect Microsoft DirectX* Frames

Got a slow spot in your game play? You don't want to know where you are spending a lot of time, you want to know where you are spending a lot of time and the frame rate is slow. VTune Amplifier XE 2013 can now automatically detect Microsoft DirectX* frames and filter results to show you what is happening in slow frames. Not using DirectX*? Just define the critical region using the API and frame analysis becomes a powerful tool for analyzing latency.

New!Better Memory Bandwidth Analysis

VTune Amplifier XE 2013 performs a more accurate memory bandwidth analysis for both reads & writes to cache and memory. It also adds bandwidth analysis for additional processor types.

Intel® Threading Building Blocks, OpenMP 4.0, Intel® Cilk™ Plus

Built-in understanding of parallel programming models means profiling data is described using familiar terms from the source, not with cryptic internal runtime labels.

New!Low Overhead Java* Profiling

Analyze Java or mixed Java and native code.  Results are mapped to the original Java source.  Unlike some Java profilers that instrument the code, VTune Amplifier XE 2013 uses low overhead statistical sampling with either a hardware or software collector.  Hardware collection has extremely low overhead because it uses the on-chip performance monitoring hardware.

New!Analyze User Tasks

The task annotation API is used to annotate your source so VTune Amplifier XE 2013 can display which tasks are executing. For example if you label the stages of your pipeline, they will be marked in the timeline and hovering will reveal details. This makes profiling data much easier to understand.

New!Tune for Intel® Xeon Phi™ Products

Hardware profiling is supported for Intel® Xeon Phi™ products and can be launched from the graphic user interface. It can collect advanced hotspots and advanced event data and has time markers for correlation of data across multiple cards. Software collection (e.g., locks and waits analysis) is not supported on Intel® Xeon Phi™ products.

New!"Hot keys" Start and Stop Analysis

Add a short cut to quickly launch performance analysis whenever you see your app running slowly.  Program hot keys to start and stop the collection of performance data.

New!Tune MPI Applications

Analyze hybrid applications using MPI and OpenMP. Install on a cluster.

New!Support for New Processors

VTune Amplifier XE 2013 is constantly adding support for the latest processors. Updates are released shortly after new processors begin shipping.

Technical Specifications

For additional information and details on new features, please see the "What's new?" articles and release notes.

 

"The new VTune™ Amplifier XE brings even more capability to an already indispensable tool. The sampling based call stack hotspots is excellent and alone is worthy of the upgrade. We have also been impressed by how the concurrency and Locks and Waits analysis can even provide useful data on complex applications such as Premiere Pro."

Rich Gerber - Engineering Manager, MediaCore, Adobe Systems Inc.

"The new interface is a joy to use. Intel® VTune Amplifier XE gives us precise, down-to-the-metal performance data that's invaluable for pinpointing hotspots and evaluating the effect of optimizations"

Daniel Schwarz, Performance Engineer, Nik Software

"Intel® VTune™ Amplifier XE's timeline is very information intensive. It organizes the data I need to tune threaded applications."

Sergey Zaritchny, Software Development Manager, Open Cascade SAS

"Last week, Intel® VTune™ Amplifier XE helped us find almost 3X performance improvement. This week it helped us improve the performance another 3X."

Claire Cates, Principal Developer, SAS Institute Inc.

"One of Intel® VTune™ Amplifier XE's best features is that it is easy to use. I did not need to read the documentation."

Richard Shepherd, Software Engineer, ESRI (UK) Limited

 

What’s New in 2013 SP1?

We continuously release new features in regular updates available to all customers with a current service agreement (one year included with purchase). Just download, install and get all the latest stuff. Here is a partial list of new features released since our first release of Intel VTune Amplifier XE 2013: (For more details, see our What’s New? summary for each update.)

More Profiling Data

  • Intel® Xeon Phi™ – memory and vectorization profiling
  • GPU for compute – Tune OpenCL. Correlate GPU and CPU activities. (Newer processors, Windows* only.)

Better Data Mining – Find Answers Faster

  • Search added to all grids
  • Timeline sorting, band height, time scale configuration
  • Loop hierarchy, overhead and spin time metrics
  • OpenMP* 4.0 scalability analysis

Easier to Use

  • Attach to a running Java process
  • Contextual help for hardware events and performance metrics
  • Easier generation of command line options from the user i/f

New OS & Processor Support

  • Intel® Xeon Phi™ coprocessor, Haswell – Windows* & Linux*
  • Windows 8 desktop and Visual Studio* 2012 & 2013
  • Latest Linux distributions

What’s New in 2013?

Here are highlights of improvements made to the 2011 product:

More Profiling Data

  • Statistical Call Counts – Make better inlining decisions
  • Hardware Events with Stacks – Lower overhead, Higher resolution
  • Uncore Event Counting – More accurate bandwidth analysis
  • Intel® Xeon Phi™ Coprocessor – Hardware event profiling

Better Data Mining – Find Answers Faster

  • Low Overhead Java* Profiling – Results map to the Java source
  • Source View for Inlined Code – (For Intel® and GCC compilers)
  • Task Annotation API – Label and visualize tasks.

Easier to Use

  • User Defined Metrics – Create meaningful metrics from events
  • Programmable Hot Keys – Quick spontaneous profiles
  • More/Better Advanced Profiles – (e.g., Bandwidth)

Videos to help you get started.

Register for future Webinars


Previously recorded Webinars:

  • Analyzing OpenCL applications with Intel® VTune™ Amplifier XE
  • Secrets of Performance Profiling – An Introduction to Intel® VTune™ Amplifier XE
  • Advanced Profiling with Intel® VTune Amplifier XE
  • Part 1: Find the bottleneck


    Part 2: Tune for Haswell (Sandy Bridge and Ivy Bridge)

  • Accelerating financial services applications using Intel® Parallel Studio XE with the Intel® Xeon Phi™ coprocessor
  • Find 3 performance scaling barriers using Intel® VTune™ Amplifier XE
  • Performance analysis on Intel® Xeon® Phi™ Coprocessor
  •  

    Download slides

  • How Intel® Parallel Studio XE is used to improve the HMMER application

Featured Articles

No Content Found

More Tech Articles

How to Use Power Collection in Intel® VTune™ Amplifier for Systems
By kevin-oleary (Intel)Posted 02/06/20130
Background Intel® System Studio is the name for the Intel’s new embedded software tool suite. This tool suite includes Intel® VTune Amplifier for Systems. This article will explain the steps you need to follow to run the new remote power collectors in VTune Amplifier for Systems on an embedded p...
How to use Intel® Inspector for Systems
By kevin-oleary (Intel)Posted 02/06/20130
Background Intel® System Studio is the new embedded software tool suite and includes Intel® Inspector for Systems. This article will explain the steps you need to follow to run Inspector for Systems on an embedded platform. Overview We will use Yocto Project* version 1.2 as an example. This pl...
How to use Intel® VTune™ Amplifier for Systems
By kevin-oleary (Intel)Posted 02/06/20130
Background  Intel® System Studio is Intel’s new embedded software tool suite.  This tool suite includes Intel® VTune Amplifier for Systems. This article will explain the steps you need to follow to run VTune Amplifier for Systems on an embedded platform. Overview The embedded OS we will be foc...
Workaround for interoperability issue with Symantec* Endpoint Client Protection software older than 12.1.2015.2015 for Intel(R) Advisor XE 2013 survey and Intel(R) VTune(TM) Amplifier XE 2013 hotspot analyses
By RAVI (Intel)Posted 01/10/20130
Problem Intel(R) Advisor XE 2013 Survey and Intel(R) VTune(TM) Amplifier XE 2013 hotspot analyses fails when running on a system with Symantec* Endpoint Client Protection software versions older than 12.1.2015.2015 (For ex: 12.1.1000.157). Symptoms As soon as analysis is started, Application Und...

Pages

Subscribe to

Supplemental Documentation

No Content Found
Subscribe to

You can reply to any of the forum topics below by clicking on the title. Please do not include private information such as your email address or product serial number in your posts. If you need to share private information with an Intel employee, they can start a private thread for you.

New topic    Search within this forum     Subscribe to this forum


frame analysis with a graphic application build with VC6.0
By Demao H.5
       hello,i am a newly user of  vtune. i am learning to develop a game  on dx9.it's IDE is VC6.0. I encoun some performan issues.  I use the vtune 2013 update 15 to do frame analysis on my application. After adding the ittnotify.h and libittnotify.lib to my application, i got two link errors :        libittnotify.lib error LNK2001: unresolved external symbol ___security_cookie        libittnotify.lib error LNK2001: unresolved external symbol @__security_check_cookie@       some document  on the internet tell me to add the bufferoverflowU.lib to my application. After adding bufferoverflowU.lib to my application , i  got another Link error :       debug information corrupt .   some document on the internet tell me that this problem is caused by the compatibility of compiler. VC6.0 is too old to use the .dll that build with newly compilers that support security_cookie or /Gs option. I consider that the reason why i encounter these problem is that the libittnotify.lib and libittnoti...
Erroneous detailed hardware metrics in VTune Amplifier XE 2013
By Peter E.6
Hello, I'm having problems with my analysis results in VTune. Whenever i do an analysis, hardware event-based metrics behave strangely. Their "aggregator" (e.g. Bad Speculation) seems to show correct values, but whenever i expand that, detailed metrics (e.g. Branch Mispredict, Machine Clears) always show zero (the blue bar is missing). To illustrate, here is a C++ code snippet that should trigger loads of L1 cache misses: /* ... */ /*kEvilOffset is 0x1000, accessing data with this offset should result in a large amount of L1D replacements (assuming a 32KB L1 data cache)*/ __declspec(noinline) void DoStuff() { for (size_t i = 0; i < kAllocSize - kEvilOffset; ++i) for (size_t j = i; j < kAllocSize; j += kEvilOffset) { data[i] *= 39; } } /* ... */After running a "General Exploitation (Sandy Bridge / Ivy Bridge / Haswell)" analysis, the blue bar for "Back-End Bound" is very wide (which - by the way - should be correct, since cache misses are in that category), but wh...
A few questions regarding hardware events
By T C13
Greetings, Looking at the below link, what exactly is the difference between non-precise and precise monitoring? Might seem a silly question- but why would anyone ever use unprecise monitoring?? http://software.intel.com/sites/products/documentation/doclib/stdxe/2013... In the below link what is the difference between a "call" and "non-call"? Would one of these refer to polymorphic virtual methods? In the same below link, I would like to see the value for the event "INDIRECT_NON_CALL" and I can see a mask. This appears to be some sort of bit manipulation device. How exactly do I see the value of this event counter? Where/what do I do with the mask? http://software.intel.com/sites/products/documentation/doclib/stdxe/2013... In short, I am trying to work out which event counters I need to look at to observe indirect branch mispredictions for virtual calls. I am presuming I am close with these "INDIRECT" event counters?
Two summary results attached- assistance with interpretation
By T C0
I have implemented some code using two approaches. I am looking at the results (attached)  and I can tell that the "faster" version had less branch mispredictions, less L1 instruction cache misses, less TLB misses but I cannot calculate how many CPU cycles were consumed. The total difference between the two designs is several billion instructions. Could somebody please glance at my results and assist me in how I can determine where the "additional" CPU cycles were consumed? These are the memory access costs I have found: L1 cache miss 4 cycles L2 cache miss 10 cycles L3 cache miss 40 - 75 cycles Hitting main memory 350 cycles TLB miss 7 - 16 cycles ucode cache miss ??? cycles The results are quite hard to interpret because they have this magic "rate" measurement- rather than actual number of CPU cycles spent for each of the metrics. I am using an Ivy Bridge CPU so I cannot see many of the event counters. Ideally I would just be able to see the raw event counters but it seems these h...
Error: Failed to allocate Injector, Error = INJECTOR_ERR_FAILED_TO_CREATE_REMOTE_THREAD
By Sophie B.11
Hello, I just installed the VTune Amplifier XD (2013) evaluation edition (I have had a license key ordered but I have not yet received it). I am trying to look into some performance issues that are occurring in some Java /or/ C++ code (the Java layer is the main codebase, but via JNI we utilize C++ code, also). The application cannot run standalone, however - so I cannot launch if from the command line; it has to run within a processing server (that runs as a java process owned by "SYSTEM", according to the Task Manager - I could be mistaken, but I do not believe that I can change this to run as Administrator, which is the user I log onto the system as). I am also running on a vCloud image, it appears that we are using VMware vCloud Director 5.1. The last point that is probably important is that it isn't the Oracle JVM, but rather our internally packaged JVM (I'm not sure if there are differences though I know that technically only Oracle JVM is supported, and I've already received ...
L1, L2 and LLC instruction and data cache misses on E5-2420 using intel vtune
By Maria M.11
Hi,  I am trying to find out the L1, L2 and LLC instruction and data cache misses of the application. I am using Intel(R) Xeon(R) CPU E5-2420 0 @ 1.90GHz. Can you please tell me how can I find instruction and data misses using intel vtune.  Secondly, if machine has hyper-threading active, should it be a good idea to turn it off to characterize the  application.   Regards,
VTune timestamp api
By Min Xu11
Hi I would like a function that would provide me the timestamp vTune uses in its timeline, so that any program trace/output can be tagged with this timestamp for a more comprehensive analysis (e.g.., I would like to know how many tasks a worker thread still have at a particular timestamp in the vtune timeline view).. This function does not appear to be available in the current vtuneapi dll.  Is it something that can be added, or can you show me now to get the equivalent value for the CPU ticks (preferred over OS ticks since more accurate)  in .NET code?   I have looked other .NET forums on accurate timestamp function but there doesn't seem to have any correlation with vTune: 1)  DateTime.Now is inaccurate (15 ms update cycle) 2)  StopWatch will drift   Thanks
hardware counter information for individual functions via commandline
By Prasanna B.9
Dear Vtune experts, I am trying to gather hardware counter information for individual functions in a code. While this seems straightforward in GUI, I haven't found a way through command line interface. In the command line, I tried two reports: (1)  amplxe-cl -R summary -r r012ge This gives me hardware counter information (instruction count, cache hit rate) for the entire code, but not for specific functions, which is what I need. (2) amplxe-cl -R hotspots -r r012ge This gives me timing information for individual functions, but no hardware counter information. Thanks  

Pages

Subscribe to Forums
  • What is the difference between "Basic Hotspots" and "Advanced Hotspots"?
  • Basic Hotspots (formerly "Hotspots")Advanced Hotspots (formerly "Lightweight Hotspots")
    Uses the software collector Uses the hardware collector and the on chip Performance Monitoring Unit (PMU)
    No driver required Requires a driver
    Runs on Intel® and compatible processors Requires a genuine Intel® processor for collection
    ~10ms resolution ~1ms resolution (finds smaller functions)
    Collects call stacks to show calling sequences New! Optional call stack collection
    Works in virtual environments Works in a virtual environment only when supported by the VM vendor (e.g., vSphere* 5.1)
  • Can I install and use Intel VTune Amplifier XE on a system with a compatible processor not manufactured by Intel® Corporation?
  • Yes. Intel VTune Amplifier XE will operate on both Intel® processors and compatible processors when analyzing applications containing Intel® instructions. Profiling features that use the software collectors (e.g., "Basic Hotspots" and "Locks & Waits") work on both Intel processors and compatible processors. Profiling features that use the hardware collectors and the on-chip performance monitoring unit (e.g., "Advanced Hotspots" and "Advanced Analysis") require a genuine Intel processor for data collection, but after collection the results can analyzed on a compatible processor.

  • Do I need to recompile?
  • No, you do not need to recompile in order to profile with Intel® VTune Amplifier XE. However, it is recommended that you have debug and symbol information available for your optimized application in order to get the most complete and useful results. Thus, your release build process may need to be modified to add symbol information to the optimized build.

  • Do I need to use the Intel compiler to use Intel® VTune Amplifier XE?
  • No, you do not need the Intel compiler to analyze applications. However, if you are using OpenMP, it is recommended that you use the Intel runtime if possible to get the best results.

  • Can I run a performance analysis on a remote system?
  • Yes.

  • Do I need multiple licenses to do remote data collections?
  • No. Once you have the product, the CLI installer (command line installer) permits the installation of collection on other systems of the same OS. You can collect the data on the remote system, but you will need a license to view the data. Copy the results directory to a system with the full product installed for viewing. For more details see “Remote Tuning Workflow” in the documentation. For installation details see “Installing Collectors on Remote Systems” in the release notes.

  • Why can’t I see my source code?
  • There are several possible reasons why VTune Amplifier XE may be unable to see your source.

    In order for source code to be visible you need to compile your code so that debug information is available. For example, on Linux*, verify you are compiling with the “-g” flag.

    You also need to let VTune Amplifier XE know where your source files, binary files and symbol files are located. To do this, open or create a Project and click on the “Project Properties” button. In the Project properties dialog, click on the “Search Directories” tab. In the pull down menu, click on “All files” and then specify the directory where your files exist. If you have any subdirectories remember to check the “Search subdirectories” box.

  • Do I need to be root to run the hardware collector used with "Advanced Hotspots" and "Advanced Analysis"?
  • No. On Linux*, you need to be root to install the driver for the hardware collector, but once it is installed root access is not required. On Linux*, depending upon the install options selected, you may need to be a member of the driver access group (“vtune ” by default) to use the hardware collector. The hardware collector is used for advanced hotspots analysis and advanced analysis. For more information see “Installing the Sampling Driver” in the documentation.

  • What file and directory permissions are required to use VTune™ Amplifier XE?
  • Because the hardware-based ("advanced") sampling analysis types require communication with the Performance Monitoring Unit (PMU) of the central processor, the installer attempts to install a device driver. For Windows*, the driver is signed and the person installing must be part of the Administrators group. On Linux*, the person installing the software must be root or have sudo access to install the driver. However, a Linux user can install the software locally without the device driver and still use the user-mode sampling analysis types: Basic Hotspots, Concurrency, and Locks-and-Waits. If the user is able to install the software as ‘root’, any user that desires to collect hardware-based samples may (depending upon the options selected during install) need to be part of the user group defined during the install. By default, this is the ‘vtune’ group, but it can be changed or omitted by accessing the Advanced options of the installer (install.sh).

  • Why can’t I import results?
  • In order to import results into VTune Amplifier XE, you must first create a project to contain the imported results. In the VTune Amplifier XE, click on the File->New->Project menu. This will bring up a dialog asking you to select a project name. Enter a name and press “OK”. VTune Amplifier XE will display the “Project Properties” dialog. If you are only importing results into the project then you will not need to specify an application name. However, if you want to view source of the imported results, you need to specify the search directories where your source and binaries are located. In the Project properties dialog, click on the “Search Directories” tab. In the pull down menu, click on “All files” and then specify the directory where your files exist. If you have any subdirectories remember to check the “Search subdirectories” box.

  • I added a path to the Search Directories, but nothing changed?
  • The Search Directories are used during finalization and that normally occurs after data collection completes. In order for new “Search directory” paths to take affect VTune Amplifier XE must re-resolve your results with the new information provided. Click on the “Analysis Type” tab and then press the “Re-resolve” button on the far right, located directly below the “Start” and “Project Properties” buttons.

  • Why do the sample counts look wrong?
  • Sometimes, the sample counts may be displayed on source lines that are not normally associated with executable code, for example, the closing brace of a ‘for’ or ‘while’ loop. This may appear to be an error but is a result of the instructions generated by the compiler. Viewing the assembly code can reveal that the debug information for the assembly instructions to which the samples are attributed, are tagged as belonging to that source line, i.e., the closing brace.

    Other times, viewing of the assembly instructions may show that certain hardware events were collected on instructions that could not possibility generate that event, e.g., a memory event on a jump instruction or an arithmetic event on a memory instruction. This is known as “event skid” and is a result of the processor being unable to stop the execution of some micro-ops before sampling the instruction pointer. Thus, the IP is pointing at a subsequent instruction by the time the sample is taken. Typically, you can determine which instruction was responsible for the event by examining the instruction flow.

  • How can I use Intel® VTune Amplifier XE to see how much time is spent doing Disk I/O?
  • If your application is doing blocking I/O, the function call attributed to the file accesses should appear in Basic Hotspots Analysis. Additionally, if you have multiple threads waiting to access a single file, the synchronization object protecting the file , e.g. a Critical Section, should show up in the Locks and Waits analysis.

Intel® VTune™ Amplifier XE 2013

Getting Started?

Click the Learn tab for guides and links that will quickly get you started.

Get Help or Advice

Search Support Articles
Forums - The best place for timely answers from our technical experts and your peers. Use it even for bug reports.
Support - For secure, web-based, engineer-to-engineer support, visit our Intel® Premier Support web site. Intel Premier Support registration is required.
Download, Registration and Licensing Help - Specific help for download, registration, and licensing questions.

Resources

Release Notes - View Release Notes online!
Documentation:
Windows* | Linux*
Documentation for other software products

Featured Support Topics

No Content Found