Earlier this week, Intel announced the release of Intel® Parallel Studio XE 2018, with updated compilers and developer tools.
Intel® Parallel Studio XE 2018 has been available in a free Beta program since April. This week’s formal release of the fully supported product is notable with new features that further enhance the tool set for accelerating HPC applications.
One of those new features worth taking a look at is Application Snapshot in Intel VtuneTM Amplifier, which gives you a quick look at an application’s performance profile to see how well, or badly, optimized it is for the latest hardware. This performance snapshot gives easy-to-understand displays of various metrics, including CPU utilization, memory operations and stalls, FPU utilization, MPI and OpenMP* parallelism, and I/O efficiency. It also gives recommendations for further analysis, and supports Intel MPI, MPICH, and Cray MPI. Application Snapshot gives immediate feedback to help determine if an application needs performance tuning.
The compilers have been updated to generate highly optimized code for Intel® Advanced Vector Extensions 512 (Intel® AVX-512) instructions for Intel® Xeon® and Xeon PhiTM processors. Intel follows programming language standards very closely, including drafts not yet official. Intel compilers support full C11, C++14, and the draft C++17 standards. Also, the suite includes Parallel STL* for parallel and vector execution of the C++ Standard Template Library.[clickToTweet tweet=”Intel® Parallel Studio XE 2018 released with updated compilers and analysis tools.” quote=”Intel® Parallel Studio XE 2018 released with updated compilers and analysis tools.”]
Both full Fortran 2008 and the draft Fortran 2015 standard are supported, as well as the Microsoft* Visual Studio* 2015 shell. The C/C++/Fortran compilers not only implement the latest OpenMP* 4.5 shared memory parallelization specifications, but many of the proposed 5.0 features as well.
Intel continues to demonstrate its commitment to scientific Python with this release of Intel Parallel Studio XE 2018. This “accelerated” Python distribution uses optimized memory management and libraries (including the Intel Math Kernel Library) to take advantage of the latest multicore and vector hardware. Up to 20x performance improvements have been noted.
The 2018 release also adds new, broader redistribution rights for Intel Performance Libraries and Intel Distribution for Python.
Intel Parallel Studio XE 2018 comes in three telescoping editions:
- Composer Edition, for building applications, includes: the C/C++ and Fortran optimizing compilers, Intel Math Kernel Library, Intel Integrated Performance Primitives library for fast image, signal, and data processing, Intel TBB C++ Threading Library, Intel DAAL data analytics library, and the optimized Intel Distribution for Python.
- Professional Edition adds analysis tools: Intel VTune Amplifier for performance profiling, Intel Inspector for memory and thread debugging, and Intel Advisor, for optimizing vectorization and thread prototyping.
- Cluster Edition adds cluster tools: Intel Cluster Checker, an expert system for analyzing cluster diagnostics, Intel Trace Analyzer and Collector for MPI tuning and analysis, and the Intel MPI Library.
Download your free 30-day trial of Intel® Parallel Studio XE 2018
