We’re pleased to announce new version of the Vectorization Assistant tool - Intel® Advisor 2017 update 1. For details about download, terms and conditions please refer to Intel® Parallel Studio 2017 program site.
Below are highlights of the new functionality in Intel Advisor 2017 update 1
Cache-aware roofline modeling: To enable this preview feature, set the environment variable ADVIXE_EXPERIMENTAL=roofline before launching the Intel Advisor.
Analysis workflow:
- Intel® Math Kernel Library (Intel® MKL) support: Intel Advisor results now show Intel MKL function calls.
- Improved FLOPs analysis performance.
- Decreased Survey analysis overhead for the Intel® Xeon Phi™ processor.
- New category for instruction mix data: compute with memory operands.
![]()
- Finalize button for “no-auto-finalize” results, when result is finalized on a separate machine.
- MPI support in the command line dialog box..
![]()
Recommendations:
- Recommendations display in Refinement Reports.
![]()
- New recommendation: Vectorize call(s) to virtual method.
- Cached recommendations in result snapshots to speed up display.
Memory analysis:
- Ability to track refinement analysis progress: You can stop collection if every site executes at least once.
Get Intel Advisor and more information
Visit the product site, where you can find videos and tutorials. Register for Intel® Parallel Studio XE 2017 to download the whole bundle, including Intel Advisor 2017 update 1.