HPC news for supercomputing professionals

Entries filed under “Tools”

News about compilers, debuggers, communications libraries, and the tools of HPC development.

New Lustres released

Joe Landman has a short post on his blog pointing to new Lustre drops. When he posted the news wasn’t reflected on the Lustre site, but it has since been updated

Lustre 2.0.0
We are pleased to announce that Lustre 2.0.0 is now GA and available for download. Lustre 2.0.0 provides support for OEL 5, RHEL 5, SLES 10 and 11 (client only), and Fedora 11 (client only). This release offers a number of significant features and enhancements, including Changelogs, Commit on Share, Lustre_rsync, and Size-on-MDS (preview). Lustre 2.0.0 supporting documentation includes the Operations Manual, Release Notes, and Change Log 2.0.

Lustre 1.8.4
Lustre 1.8.4 is now GA and available for download. Lustre 1.8.4 provides support for OEL 5, RHEL 5 and SLES 10 and

Also posted in System Management | 2 Comments

Microsoft parallel runtime expected to go commercial next year

ZDNet wrote last week about a new HPC offering that Microsoft is evidently planning on moving from research to commercial product in the coming months. The platform is called “Dryad”

Dryad is an ongoing Microsoft Research project dedicated to developing ways to write parallel and distributed programs that can scale from small clusters to large datacenters. There’s a DryadLINQ compiler and runtime that is related to the project. Microsoft released builds of Dryad and DryadLINQ code to academics for noncommercial use in the summer 2009.

As you can see from the diagram (click for a larger view), there is a lot of technology in the platform, including a compiler, runtime, a new file system …

Also posted in Computing Research, Datacenter operations, HPC Software, System Management | 6 Comments

AMD revs Stream SDK, now with OpenCL 1.1

Last week AMD announced that is has released rev 2.2 of its Stream SDK. In case you need a refresher

ATI Stream technology is a set of advanced hardware and software technologies that enable AMD graphics processing cores (GPU), working in concert with the system’s x86 cores (CPU), to accelerate many applications beyond just graphics. This enables better balanced platforms capable of running demanding computing tasks faster than ever, and sets software developers on the path to optimize for the upcoming AMD Fusion Family of Accelerated Processing Units (APUs).

This version offers “full support” for OpenCL 1.1, which was ratified in June.

“Availability of the ATI Stream SDK v2.2 with OpenCL 1.1 support is a great example of how

Also posted in GPUs, HPC Hardware | 3 Comments

Vuduc wins NSF CAREER Award to make HPC better “by any means necessary”

In early June the NSF announced that Georgia Tech’s Richard Vuduc received an NSF CAREER Award for his work in tuning software to run on parallel systems. The name of his proposal, “Autotuning foundations for exascale systems”, attracted my attention and Rich agreed to tell us a little about himself, his work, and this prestigious award.

Here’s an excerpt from the interview where Dr. Vuduc describes the technical work in his proposal:

The proposal has two major research thrusts, one that explores analytical and statistical performance models to guide tuning, and another that explores tuning in emerging dataflow-like programming models. In both cases, we want methods that work on (a) the kinds of sparse, irregular, adaptive computations that I’ve been studying for some time now and that

Also posted in Computing Research, Featured Stories, HPC Education and Training, HPC People, HPC Software | 1 Comment

NAG grows C and C++ libraries

Late last week NAG announced it has expanded its NAG C library with 150 new functions, bringing the new total to over 1300.

The new release of The NAG Library includes two new chapters on wavelet transforms and global optimization. A new sub-chapter has also been introduced on option pricing. Enhancements have been made in the areas of statistics, optimization, linear algebra, ordinary differential equations, regression, random numbers, sorting, and special functions.

More at the company’s web site.

Also posted in HPC Software | 1 Comment

NVIDIA’s CUDA dev tool reaches GA

NVIDIA announced this week that its Nsight plugin for Microsoft Visual Studio has reached production readiness after entering beta back in April.

High performance computing (HPC) developers in fields such as oil and gas, bioscience and finance, are increasingly tapping the massively parallel processing capabilities of GPUs and Windows HPC Server 2008 to achieve significant performance increases in their work. NVIDIA Parallel Nsight connects these worlds, providing the professional tools required to develop and deploy HPC cluster applications.

“Research shows that developers believe the most difficult tasks when developing parallel applications are debugging, performance tuning and designing parallel algorithms,” said David Rich, director, technical computing at Microsoft Corp. “By integrating GPU computing into Visual Studio, NVIDIA’s Parallel Nsight is

Also posted in GPUs, HPC Hardware, HPC Software | 1 Comment

Digipede Network 2.4 release

Digipede is one of those companies quietly working to make HPC easier for the vast majority of the world that doesn’t currently, and doesn’t want to, use anything other than Windows. They launched at DEMO in 2005 but we haven’t written about them in a while, so here’s a refresher if they are new to you (by the way, if you’d like us to run a 411 on your company, send me an email ).

Their latest news is about the launch of version 2.4 of their flagship product, Digipede Network. In a nutshell, by automatically deploying .NET assemblies (and related files), then distributing and executing .NET objects natively, the Digipede Network adds support for high-performance .NET applications to Windows HPC Server.

Also posted in Enterprise HPC, HPC Software | 1 Comment

Live chat with CUDA inventor Ian Buck

NVIDIA is hosting a series of live text chats this summer where they’ll host high profile GPU personalities (users, developers, etc.) with the opportunity for fellow travelers in the GPU community to interact with them. Each chat will last 30 minutes, and is open to anyone.

The first guest is Ian Buck, inventor of the Brook project and CUDA, and currently Software Director of GPU Computing at NVIDIA. Ian will preview his talk at the GPU Technology Conference (GTC) later this year, and answer questions about the future of GPU and CUDA computing.

From the website for today’s chat

Hey there, my name is Ian Buck and I’m the Software Director of GPU Computing here at NVIDIA. I helped start the CUDA

Also posted in Events, GPUs, HPC Hardware | 2 Comments

PGI releases Fortran for Visual Studio

Today The Portland Group announced PGI Visual Fortran for Visual Studio 2010.

“With this latest release of PVF, PGI Fortran compilers and tools for multi-core processors and GPUs are available through Visual Studio 2010 to the large base of scientists and engineers developing for Windows,” said Douglas Miles, director, The Portland Group. “PVF’s world-class performance and state-of-the art compiler technologies allow developers to leverage the wide array of new microprocessor and accelerator innovations coming out of Intel, AMD and NVIDIA together with the productivity advantages of Microsoft HPC technologies.”

PGI Visual Fortran is based on PGI’s native OpenMP and auto-parallelizing compiler for the Fortran 95/2003 programming languages.

Key features include:

  • Native Fortran 95/2003 Visual Studio project system;
  • Fortran text editor extensions, intrinsic function

1 Comment

NSF to formalize technology adoption as it begins TeraGrid shutdown

Recently NCSA announced that it is part of a partnership including TACC, PSC, and the NICS that has received an $8.9 million, five-year award to run the eXtreme Digital (XD) Technology Insertion Service (TIS).

John Towns, NCSA’s director of persistent infrastructure, TeraGrid Forum chair and principal investigator (PI) on the Xtreme Science and Engineering Discovery Environment (XSEDE) TIS grant, gave me a little more detail by email on what they are actually going to be up to

The Technology Insertion Service  will evaluate technologies that could enhance the productivity of researchers who use of TeraGrid and who will use its successor, eXtreme Digital.  This includes a broad range of “traditional” technologies — community applications, middleware, compilers, libraries, systems software,

4 Comments

Advertisement

NVIDIA Fermi

insideHPC.com is a production of indigoBit, LLC. © 2006-2010 Sitemap