HPC news for supercomputing professionals

Entries filed under “HPC Hardware”

Hardware news and announcements in technologies related to HPC.

Eurotech Launches Aurora AU-5600

Eurotech, today, announced the launch of the latest in the Aurora series of supercomputing platforms.  Deemed a “Green Petascale computing platform”, the Aurora AU-5600 the latest Intel 5600 series “Westmere” processors.

Aurora has always been a clear leader in terms of capability, TCO and optimal resource use” said Giampietro Tecchiolli, VP and CTO of the Eurotech Group. “The new 6-core Intel Xeon 5600 series are a perfect

Also posted in Compute, HPC | 1 Comment

UV login seen in the wild

During his presentation today at the Newport HPC conference Eng Lim Goh, SGI’s CTO and the brains behind SGI’s x86-based shared memory future, logged into an SGI UV and gave it a little exercise for the audience. He mentioned that it was 1,000 cores in one 2 TB shared memory space. Was good to see things are on track to start shipping in summer.

Also posted in Compute, InsideTrack | 1 Comment

Yotta Yotta IP features in Gelsinger’s plan from EMC

Stacey Higginbotham at GigaOm is reporting this week on Pat Gelsinger’s vision (fresh off 30 years at Intel) for EMC’s to virtualize and federate storage on a global scale

Pat Gelsinger, who moved to EMC late last year after 30 years at Intel, is stirring things up at the storage giant with a plan to virtualize and federate storage so data and compute can truly be linked together…Gelsinger’s answer is caching. Imagine a two-way

Also posted in Cloud HPC, Storage | 1 Comment

SGI Announces Origin 400 Workgroup Blade Platform [UPDATED]

Last week, John West had a peek at a new SGI product via an accidental RSS headline.  Well, he was right.  SGI today announced the official release of the Octane Origin 400 platform.  According to the release, its “an integrated workgroup blade system that features compute and storage area network (SAN) storage functionality, making it ideal for remote …

Also posted in Compute, Datacenter operations | 1 Comment

Landman’s list of file systems to watch

Joe Landman has a post up this week noting some of his experiences with “new” filesystems worth watching. He talks about GlusterFS and Lustre, and then the newer ceph, Btrfs, tahoe-LAFS, and a few others.

But what about the “new” systems?

First, there is ceph. Ceph is a distributed object store done right. We have set up a few test systems with it, and will get more aggressively into it later this year, including (likely)

Also posted in Storage | 1 Comment

C-Dac Reaching for a Petaflop by 2012

Pune, India-based Centre for Development of Advanced Computing (C-Dac) is designing and developing some new technology for their upcoming PARAM supercomputer.  The new machine will be ready by 2012 and is expected to break the Petaflop barrier.

he total investment to design and develop a brand new high speed petaflop PARAM supercomputer will be over Rs 1,000 crore,” Rajan T Joseph, director general, C-Dac, told Financial Chronicle. The cost of its maintenance alone

Also posted in Compute, Computing Research, Datacenter operations, HPC | 2 Comments

Russian University Shooting for Petaflop-Scale System

The chairman of the state parliament of the Russian Federation announced last Friday that the country’s biggest university would obtain a supercomputer that breaks a Petaflop in performance.  So who’s the lucky recipient?  According to the source article, the candidate recipient will be the Moscow State University.

The study regarding techno-economical efficiency of a PetaFLOPS supercomputer based on the ‘Lomonosov’ supercomputer is present. I will put this matter on consideration of the president and

Also posted in Compute, HPC, New Installations | 1 Comment

Cray CX-1 Support for Intel Westmere

Cray announced today that their CX-1 line of deskside supercomputers will now support the latest Intel Xeon 5600 series of silicon.  Formerly codenamed ‘Westmere’, the latest in silicon goodness from Intel is based on the existing 32nm manufacturing process, but includes a new microarchitecture.  The new microarchitecture features two more compute cores per socket (from four core to six core) while maintaining the same thermal …

Also posted in Compute, HPC | 1 Comment

Voltaire intros midrange 40 Gbps switch plus new software to speed up MPI collectives

This week network equipment manufacturer Voltaire announced two new pieces of its network gear product line. First up is the company’s new midrange InfiniBand switch, aimed at cloud computing and enterprise HPC applications

Voltaire…introduced a mid-size 40 Gb/s InfiniBand director switch expanding its full line of 40 Gb/s platforms.  The Voltaire Grid Director 4200 is a high performance, ultra-low latency InfiniBand switch that is ideal for high performance

Also posted in Network | 1 Comment

NEC Builds New Cluster in Germany For Astro and Geophysics

ZDNet DE posted an article today detailing a new joint venture supercomputing project between the Society for scientific computing Göttingen (IT Center), the Max Planck Institute for Solar System Research (MPS) and the Institute of Geophysics, Georg-August University in Göttingen. The new machine was built using NEC LX-2200 nodes, anchored by Intel Nehalem processors. The overall system consists of nearly 200 HPC …

Also posted in Compute, HPC, New Installations | 4 Comments

Advertisement

StarNet Communications

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