Intel® VTune™ Amplifier XE 2013
Intel® VTune™ Amplifier XE is an easy to use performance and thread profiler for C, C++, C#, Fortran, Java and MPI developers. No special recompiles are needed, just start profiling. Hotspots are highlighted on the source. A powerful timeline makes it easy to tune your application and scale performance on multicore processors.
New for Update 14!
- Optional unless you need new processor or feature support:
- Intel® microarchitecture code named Silvermont support
- Intel® Xeon® E5-2600 v2 & E5-1600 v2 processors based on the Intel microarchitecture code name IvyBridge-EP support
- GPU OpenCL™ kernel analysis extended with memory transfers, visualization of OpenCL API and computing queue (Windows* only)
- Ability to cache source files and explore collected performance statistics later even if the source file has been changed
- Standalone interface improved to provide more workspace for the analysis results
- Ability to show kernel stacks as continuation of user stacks
- Simplified syntax for searching binary and symbol files with the –search-dir and source-search-dir command line options
- Event-based stack sampling analysis of system processes for kernels and drivers (Windows* only)
Note: We are now labeling analysis tool updates as "Recommended for all users" or "Optional unless you need…". Recommended updates will be available about once a quarter for users who do not want to update frequently. Optional updates may be released more frequently, providing access to new processor support, new features, and critical fixes.
Installer for Intel® Vtune™ Amplifier XE 2013 Update 14 for Linux*
Installer for Intel® Vtune™ Amplifier XE 2013 Update 14 for Windows*
* Other names and brands may be claimed as the property of others.
Microsoft, Windows, Visual Studio, Visual C++, and the Windows logo are trademarks, or registered trademarks of Microsoft Corporation in the United States and/or other countries.
Next: What's New in Update 13?
If your application uses OpenCL software technology and is doing substantial computation work on the GPU, you may capture the timing (and other information) of OpenCL kernels running on Intel HD Graphics by enabling the Trace OpenCL kernels on Processor Graphics option during analysis configuration. To view information about all OpenCL kernels running on the GPU, in the Graphics window switch Grouping to Computing Task Purpose / Computing Task (GPU) / Instance. VTune Amplifier identifies the following computing task purposes: Compute (kernels), Transfer (OpenCL routines responsible for transferring data from the host to a GPU), and Synchronization (for example, clEnqueueBarrierWithWaitList). The column “Data Transferred” representing all the data “transferred” with average bandwidth:
To view details on OpenCL kernels submission, in particular distinguish the order of submission and execution, and analyze the time spent in the queue, zoom in and explore the Computing Queue data in the Timeline pane. You can click a kernel task to highlight the whole queue to the execution displayed at the top layer:
Synchronization tasks are marked with vertical hatching . Data transfers are marked with cross-diagonal hatching .
For more details please refer to the “Analyzing Applications Using Intel® HD Graphics” and “Interpreting GPU OpenCL™ Application Analysis Data” topics in the product help.
In the VTune Amplifier XE Update 14 standalone interface menu and toolbar layout was improved to provide more vertical space while exploring analysis results. Notice that Menu is now invoked by the button at the top right corner, use it to control result collection, define and view project properties, and set various options:
For more details on the user interface controls please refer to the “Standalone VTune Amplifier Interface” topic in the product help.
Save your source files in the cache. You can go back to the cached sources at any time in the future and explore the performance data collected per code line at that moment of time. To enable the option go to Menu > Options > Intel VTune Amplifier XE 2013 > Source/Assembly and check Cache source files check box. Then VTune Amplifier caches your sources in the result database when you open the Source window for the first time and provides the following message:
When you open the Source window for this result for the second time and if the source file has been changed, the VTune Amplifier opens the source from the cached file with the proper notification. For more details please refer to the “Pane: Options - Source/Assembly” topic in the product help.
You can use the VTune Amplifier to profile the Windows kernel-mode process and analyze all privileged resource operations (for example, memory management, paging) it is responsible for or to explore your multithreaded kernel-mode drivers running in the context of this process. If you are a driver developer, this option can help you profile asynchronous driver threads and identify system resource utilization issues (for example, issues caused by frequent page allocations). To analyze the system process, run the VTune Amplifier with administrative privileges and configure the analysis target to attach to PID 4. For more details please refer to the “Attaching to a Process” topic in the product help.
To view kernel stacks in the user functions stacks select the User/system functions call stack mode on the filter toolbar:
To locate the call of the kernel function in the assembly code, double click the function in the Call Stack pane:
With the VTune Amplifier XE 2013 Update 14 you may perform hardware event-based sampling analysis on Intel(R) microarchitecture code named Silvermont by using Advanced hotspots from Algorithm Analysis tree and General Exploration from Intel Atom Processor Analysis, or by creating a new custom Hardware Event-Based Sampling Analysis.
With the VTune Amplifier XE 2013 Update 14 you may perform hardware event-based sampling analysis on Intel(R) microarchitecture code named IvyBridge-EP by using Advanced hotspots from Algorithm Analysis tree or General Exploration and Bandwidth from Sandy Bridge/Ivy Bridge/Haswell Analysis tee.
When finalizing the collected data and generating reports, the Intel® VTune™ Amplifier searches supporting user files to display analysis information in relation to your source code. For proper resolving symbol information, use -search-dir action-option to specify directories that should be searched for binary (executables and dynamic libraries) and symbol files (typically .pdb files). To enable the source code view in the command line report use -source-search-dir option for searching source files.
For more details please refer to the “Specifying Search Directories” topic in the product help.