Multicore.info pointed me at an ejoyable rant at pcPRO on NVIDIA’s recent comments about Larrabee and ATI. The article starts out with Andy Keane, general manager of the company’s GPU computing group, banging on the hype Larrabee’s getting even though so much is “undefined.” Ok, fair enough. Not a lot of details. Then he says […]
Archives for August 2008
"Performance is not the problem," a vote against multicore
August 25, 2008 by
I like contrarians. Probably because I’m usually too lazy to avoid group think myself, and I’m glad to be saved the effort. Jonathan Edwards, a Research Fellow at MIT, isn’t excited about the move to multicore More importantly, I believe the whole movement is misguided. Remember that we already know how to exploit multicore processors: […]
MetaRAM's technology to triple amount of RAM you can attach to a processor
August 25, 2008 by
From Ashlee Vance at The Register on Friday comes news of MetaRAM’s latest developments, which are set to lead to single Intel-based servers sporting between 144 and 288 GB of memory: MetaRAM is led by Fred Weber, the former CTO at AMD. The company launched in February with its unique brand of memory stuffing technology. […]
OptimaNumerics ports numerical libraries to Tesla
August 22, 2008 by
OptimaNumerics announced yesterday that their libraries now support NVIDIA’s Tesla platform With OptimaNumerics Libraries, applications written in high level languages such as C and Fortran, will be able to transparently access NVIDIA GPUs. This combination of the high performance of NVIDIA Tesla GPU platform with OptimaNumerics Libraries will enable software developers to save development time […]
NVISION 08 to feature CUDA developer's conference
August 22, 2008 by
Here’s a quickie The NVISION 08 visual computing conference, being held in San Jose, CA on 25-27 August, includes a CUDA Developer Conference for software developers interested in accelerating their applications using the NVIDIA® CUDA™ software development environment, which is based on the industry-standard C programming language. The event will include users talking about CUDA […]
Lincoln Labs: largest single problem ever run
August 22, 2008 by
Jeremy Kepner emailed yesterday to share some exciting news from MIT’s Lincoln Labs’ TX-2500, part of the LLGrid interactive/on-demand parallel computing system. MIT Lincoln Laboratory (www.ll.mit.edu) has demonstrated a 0.5 Petabyte calculation on the HPC Challenge Stream benchmark (www.hpcchallenge.org) using its TX-2500 computer and pMatlabXVM software. This is the largest single problem ever run on […]
PGI offers Fortran compiler to students for free
August 22, 2008 by
PGI emailed me yesterday to share news that they’ve …announced the Free PVF for Education Program. Now through 30 June 2009, PGI is offering a single-user node-locked license to PGI Visual Fortran at no cost. To qualify, an applicant must be a registered student or faculty member of an academic institution and currently living in […]
CUDA 2.0 out today
August 22, 2008 by
NVIDIA emailed me today to let me know about an announcement that’s not up on their web site yet: NVIDIA Corporation has announced the production release of CUDA 2.0, the latest version of its award-winning C language programming environment for GPUs that enables software developers to tap into the massively parallel architecture of the GPU […]
DataRush RC1
August 22, 2008 by
Back in March I wrote a feature about Pervasive’s DataRush for HPCwire. The article brought together some things I heard at the Rhode Island conference along with some of the data growth figures I’d been hearing. Pervasive Software is one of the companies working on the software front of the data intensive computing space, developing […]
Amazon Adds Persistent Storage to EC2
August 21, 2008 by
Amazon, today, announced the launch of a new web service today called the Elastic Block Store, or EBS. The latest service provides persistent storage for EC2 computing instances. Previously, EC2 customers were only able to access temporary storage as part of the compute instance. [Think scratch storage]. Namely via the S3 online storage offering. The […]
