News

Vc: Vc 0.6.0 Released

Added by Matthias Kretz over 5 years ago

I'm pleased to announce the release of version 0.6 of the SIMD vector classes for C++.

The major improvement of 0.6 over 0.5 is its support for the AVX instructions in current AMD and Intel CPUs. For example, the Mandelbrot benchmark shows a two-fold speedup of AVX over SSE, and there were no code changes required in the Mandelbrot code itself.

There were also many other small changes to improve compiler compatibility, simplify API usage and optimize resulting binary code.

Download:
Vc-0.6.0.tar.gz

Vc: Vc 0.5.0 Released

Added by Matthias Kretz over 5 years ago

The Vc 0.5 branch was stable for a long time, already. So I finally tagged 0.5.0.
Nevertheless, for new projects I encourage to start with the 0.6 series already.

Download:
Vc-0.5.0.tar.gz

Vc: Vc 0.6 Alpha1 Released

Added by Matthias Kretz almost 6 years ago

Beginning the process of stabilizing for 0.6. The major change for this version is the introduction of an AVX implementation.

Please report any bugs or problems to the mailing list.

Download:
Vc-0.5.80.tar.gz

Vc: Vc 0.5 RC1 Released

Added by Matthias Kretz almost 6 years ago

0.5 really seems stable enough. I'll tag 0.5 soon unless I get a bug report.

Download:
Vc-0.4.91.tar.gz

High Performance Linpack: Version 1.1.0 Release

Added by Matthias Bach over 6 years ago

We are pleased to announce the release of [[110|HPL-GPU 1.1.0]]. This version includes two important new features:

  • Multi-GPU support — In combination with CALDGEMM 1.1.0 you can now use multiple GPUs per host
  • Heterogeneous system support — You can now utilize systems of different speed efficiently

These features are, however, still to be seen as experimental. Due to the limited of available multi-GPU systems and heterogeneous clusters with AMD GPUs they have not yet been tested extensively. If you experience problems in your specific combination feel free to contact the authors by joining ##caldgemm on irc.freenode.net or via the CALDGEMM mailing list.

You can download the tarball from the files section or get the source via tag v1.1.0 from the [[git repository]]. Installation instructions can be found in the [[wiki]]. If you are interested in some more in depth information you should have a look at the Technical Report that was published along with CALDGEMM 1.0.0.

CALDGEMM: Version 1.1.0 Release

Added by Matthias Bach over 6 years ago

We are happy to release version 1.1.0 of CALDGEMM. This is the first version of CALDGEMM with support for multiple GPUs. In addition it contains some general bugfixes, e.g. for a deadlock in GotoBLAS on systems with less then four cores. You can get the tarball from the files section of the website or clone the [[git repository]].

CALDGEMM allows you to use AMD GPUs to achieve the highest DGEMM performance possible on current hardware, basically reaching peak performance. It utilizes GotoBLAS for the CPU side of combined CPU/GPU computation.

Vc: Vc 0.5 Beta 1 Released

Added by Matthias Kretz over 6 years ago

The 0.5 branch has stabilized to a point where I don't expect any bigger issues to come up anymore. It is now tested on several different vanilla distribution installations - and shows no issues there.

Download:
Vc-0.4.90.tar.gz

NUMA Bench: NUMA Bench 0.9.2 released

Added by Matthias Kretz over 6 years ago

0.9.1 did not link against the shared libnuma. This prompted a quick new release.

This release adds nicer PDF output and makes static builds easier.

The tarball and a static binary can be downloaded in the Files section.

NUMA Bench: NUMA Bench 0.9.1 released

Added by Matthias Kretz over 6 years ago

This is the second release of NUMA Bench which fixes two issues:
  • The cpuid implementation was out-of-date and didn't recognize the cache sizes of newer Intel CPUs anymore. Thus the benchmark left out a lot of tests.
  • The NUMA layout was hardcoded previously, i.e. 24 cores, and 4 NUMA nodes of 6 cores each. Now it is possible to compile against libnuma and let libnuma figure this out correctly.

The tarball and a static binary can be downloaded in the Files section.

NUMA Bench: NUMA Bench 0.9 released

Added by Matthias Kretz over 6 years ago

This is the first standalone release of NUMA Bench. It is the same code as was previously available in the membench_single branch of Vc. This package is much reduced in size to only serve the one purpose of providing the benchmark.

The tarball and a static binary can be downloaded in the Files section.

« Previous 1 2 3 Next »

Also available in: Atom