insideHPC.com

Internet photo printer Shutterfly announced in April that it’s added 2.5PB of storage from DataDirect to hold the pictures you upload for printing.

From the release, DataDirect announced

that it has delivered and installed multiple S2A9550 storage solutions equal to 2.5 PB, to Shutterfly, Inc. an Internet-based social expression and personal publishing service. DataDirect’s S2A9550 is the industry’s highest performing and most

…

I missed this when it happened last week, and in case you did too, the Grid Computing Centre Karlsruhe (Germany) just agreed to install a 4,500 terabyte storage solution from NEC. According to the release

NEC has already installed 500 terabytes in disk arrays (SATA and SCSI) with 28 dual 4 Gbit/s fibre channel controllers and 45 dual Opteron servers.

…

Univa Corporation announced availability of Univa Globus Cluster Edition 2.0, a cluster software solution for high performance computing.

According to the release (inconveniently available as PDF only), the software incorporates widely-used open source software including Sun’s Grid Engine, Ganglia, Rocks, and Globus.

Hey, we’ve got stuff. Well, things. That’s right, things!

I’ve added another wallpaper, “My other computer is super.” I think it’s the cat’s pajamas. I’m experimenting with this one on a t-shirt, and there are buttons in the works, too. I tell you, I don’t know when I’ve seen more exciting swag from an HPC news site.

Download a wallpaper now …

SeekingAlpha today covered Rackable’s troubling first quarter financial report:

Rackable Systems swung to a $10.2 million ($0.36/share) loss, from a $0.23/share profit a year ago, blaming pricing pressures on its three top accounts and surging expenses. Revenue dropped 15% to $72 million.

The problems were with poaching of Rackable’s marquee customers: Amazon, Yahoo!, and Microsoft. According to coverage at The …

Charlie Catlett has an interesting post on the TeraGrid Whiteboard about a new system developed at Argonne to support urgent computing:

Pete [Beckman, at Argonne National Lab] and his team have developed an interesting capability called Special Priority and Urgent Computing Environments (SPRUCE) to give resource providers a set of tools to implement local policy for dealing with high-urgency jobs.

This

…

Nick Carr (whom I’ve never met but insist on calling “Nick” like we know each other because it makes me look like one of the cool kids) has a post with an interesting factoid.

The article is about two new special-purpose enterprise supercomputers. The first is IBM’s new Cell-processor-sporting mainframe, designed as “a server system capable of permitting hundreds of …

HP has extended its entry-level HPC offering (accessible through its web configurator) with its Cluster Platform Express system. The Express systems, which focus on single-rack solutions that are quick to ship and easy to integrate, feature HP’s new c-Class blades. The clusters can be configured with up to 48 Xeon- or Opteron-based compute nodes, and either GigE or Infiniband …

Liquid Computing is announcing today its first customer acceptance of the LiquidIQ platform. The system was placed at the Army High Performance Computing Research Center (AHPCRC). No details in the release on the configuration (if you’re a reader with details, let me know).

From Liquid’s web site:

Liquid Computing Inc. is first to deliver a new class of computer system

…

Reader Jay Blair pointed me to this story at DaniWeb

IBM has today announced the availability of an open beta version of its virtual Linux environment to enable x86 Linux applications to run without modification on POWER processor-based IBM System p servers. Designed to reduce power, cooling and space by consolidating x86 Linux workloads on System p servers, it will

…

Story in the Inquirer (pointer courtesy Sun’s HPC Watercooler) on China’s hypercompetitive technology thrust:

While other regional HPC powers, Japan and Korea, aim for one Petaflop+ class system each in the next few years, China is pushing to the forefront of the game by going for at least two, most probably three, Petaflop (sic) supercomputers by 2010.

Researchers from the University of Texas (Austin) plan to show a motherboard featuring four new TRIPS (Tera-op Reliable Intelligently adaptive Processing System) processors next week. From coverage at The Register:

“The processor core is composed of multiple copies of five different types of tiles interconnected via microarchitectural networks,” UT says on its website. “Each core may be configured in a

…

NVIDIA announced this week its new Quadro FX 5600 GPU. At $17,500 you probably won’t be putting any of these in that cluster of old PCs you have in the basement at home.

It’s a very beefy card, with a 3 GB frame buffer, support for the CUDA GPGPU programming API, and enough graphics power to push up to 80-billion pixels …

Dan Reed has a post on his blog talking about the big points in PCAST’s successor to the 1999 PITAC report produced under the chairmanship of Kennedy and Joy.

Professor Reed draws out this thread (in red) as the main take home: the leadership position of the U.S. in IT is at risk. Their report recommends four actions:

Expand the US …

IBM and The Nature Conservancy announced today they are joining forces to enlist HPC technologies to save some of the world’s great rivers.

From the release

Working through The Nature Conservancy’s Great Rivers Partnership, the two organizations will build a new computer-modeling framework that will allow users to simulate the behavior of river basins around the world, helping inform policy and

…