Today (well, actually it’s yesterday now as I write this, but I’m still up so I’m taking credit for the same day post) Aaron Tersteeg sent me an email updating me on a blog post that Intel’s James Reinders posted to the Intel software blog. The company has revved a bunch of it’s development and cluster tools
Today we released updates for our C++ and Fortran compilers, our Intel Math Kernel (MKL) and Intel Integrated Performance Primitives (IPP) libraries and Cluster toolkits. Noteworthy additions include outstanding performance enhancements, support of Intel Advanced Vector Extensions (AVX) and inclusion of some elements that debuted in Intel Parallel Studio last month.
There is a lot of good stuff in this post, which I commend to your reading list. Of interest to me was the performance improvements that some users are seeing
If you’ve not moved from the 10.x to 11.x compilers, you will want to consider doing that. Aside from new functionality such as parallel debugging, OpenMP 3.0 and AVX support – you are very likely to see pleasing performance boost esp. on the latest Intel and compatible processors. Several customer have told us of 10% performance gains in moving from 10.x to 11.1.
Along with gains of up to 45% in certain routines of the Intel Math Kernel library, and
Of course, Intel Integrated Performance Primitives (IPP) and Intel MPI library have similar success stories – and you will want to stay up-to-date for the latest performance. With IPP 6.1, task parallelism usage gains give as much as 250% multicore performance scaling while the PNG codec added to Unified Image Codec framework offers 300% faster encoding than the open source reference version. Intel MPI 3.2.1 offers industry leading performance with low latency and high bandwidths, and now uses direct inter-process memory copy for increased bandwidth on Windows systems.
James also comments in his post that a lot of the questions he got around the Parallel Studio launch last month (which I covered here) where related to when all that goodness would come to Linux — I can vouch for this, since I asked those questions myself when I interviewed him. The good news is that the migration has already begun
The Intel Parallel Debugger Extensions have been added to Intel C++ Compiler, Professional Edition for Windows. This allow serializing parallel regions, finding data sharing violations, breaking on re-entrant functions, viewing all active thread structures, OpenMP* task teams and trees, barriers, locks, and waits. Of course, this works in current versions of Visual Studio (both 2005 and 2008).
The Intel C++ Compiler 11.1 offers all the functionality of Intel Parallel Composer, plus the AVX support, on Windows, Linux and Mac OS X. We’ve added Eclipse CDT 5.0 support, SLES11 support, Native Intel 64 compiler for Mac OS X, and we support the new Mac Xcode IDE ability to relocate the tools installation directories.
There is a lot more in the full post. Check it out.