Archives for October 2009

New Chinese CUDA centers of excellence

The Google nets turned up this announcement today that NVIDIA has designated two institutions in China as CUDA Centers of Excellence NVIDIA today announced that the Institute of Process Engineering (IPE) at the Chinese Academy of Sciences (CAS) and Tsinghua University have been recognized as CUDA Centers of Excellence for their commitment to furthering GPU […]

Asus launches new 'personal supercomputer'

Several places are reporting (Engadget and Computerworld, for example) that Asustek, the company that makes motherboards and the better known Eee PC has launched a personal supercomputer around NVIDIA’s GPUs. As The Register points out the specs that have been reported don’t make sense, since with 3 Teslas it should be good for nearly 3 […]

NSF adds another .75M to Blue Waters pot for user jump start

There are lots of good things to say about the way the Blue Waters project is being handled, from the stream of news and videos keeping the community in the loop on the build (like this one), to the upfront engagement NCSA is fostering with potential users. Along that last front the NSF announced this […]

Voltaire Announces Revenue Growth

Voltaire has announced that it has grown revenue by 35% from Q2 of 2009.  Revenues for Q3 of this year hit $14.5 million, compared to $14.7 million in the third quarter of 2008.  This was up 35% from the $10.7 million of revenues in Q2 of 2009.  The company discussed more of the details regarding […]

Cray to Announce Q3-2009 Results

Cray announced this morning that they will release their third quarter, 2009 financial results to the public this Thursday, October 29th.  As always, they will do so during an investor conference call at 1:30PM PST.  For more info on how to dial in, read their full release here.

Moving beyond Pax MPI into the Exascale

As part of our series with some of the big thinkers in HPC today, insideHPC talked with Thomas Sterling about execution models and their role in the next generation of exascale computing.

It is Sterling’s take that exascale computers will not be a refinement of where we are today. The hardware, system software, and application infrastructure needed to support billion-way parallelism is a fundamental departure from the architecture curve we have been on to date, and we need to step back and critically assess where we want to go before we strike off in whatever direction seems most solicitous at the time. We probably can get to the exascale without doing a lot of heavy lifting on the execution model upfront, but with so much at stake, why take the chance?

NVIDIA Posts Recordings of GTC Keynotes and General Sessions

NVIDIA has posted a growing series of videos from the various keynote speeches and general sessions of the GPU Tech Conference back earlier this month. Of special note is Jen-Hsun Huang’s keynote speech and the Breakthrough’s in High Performance Computing general session.  The list of recorded sessions being posted continues to grow, so check back […]

S3 Graphics Launches Chrome 5400E with OpenCL Support

S3 Graphics has announced the launch of its Chrome 5400E embedded graphics processor.  We usually feature graphics articles specifically mentioning NVIDIA and/or ATi at insideHPC.  However, the Chrome 5400E card from S3 features a native OpenCL 1.0 engine.  OpenCL is an open, cross-platform standard for using the power of GPUs and other accelerator-class devices to […]

Acceleware Delivers Enhanced Processing Capabilities for RTM

Acceleware announced today a new version of AxRTM for Reverse Time Migration.  This release allows seismic data users of any use to process RTM on existing x86 clustered platforms or within heterogeneous computing environments using GPUs.  They have also added Tilted Transverse Isotropy [TTI] support in this release of AxRTM.  TTI provides enhanced accuracy for […]

NVIDIA and VSG Accelerate Oil & Gas Exploration

NVIDIA and Visualization Sciences Group [VSG], today, announced the latest release of the Open Inventor 3D Graphics toolkit.  The latest release will employ NVIDIA’s CompleX scene-scaling acceleration engine, which enables the visualization and manipulation of huge data sets.  Oil and Gas data sets most certainly fit that description.  Release 8.1 of Open Inventor allows 3D […]