Since 1986 - Covering the Fastest Computers in the World and the People Who Run Them

Language Flags
June 22, 2007

The Week in Review

by John E. West

Here's a collection of highlights, selected totally subjectively, from this week's HPC news stream as reported at and HPCwire.

>>10 words and a link

IMSL Fortran library now works with Intel Visual Fortran for Windows;

Nature: Delft researchers carry out calculations with two quantum bits;

36k-CPU Blue Gene comes online, “New York Blue” peaks at 100 TFLOPS;

Norway commits to 50 TFLOPS Cray XT4;

CRA names first leaders of Computing Community Consortium;

Austria installs fastest vector super for weather;

Sun's Blackbox finds its first customer home, painted white;

>>Bull's PR push: can HPC SuperBowl commercials be far behind?
I missed it, but Michael Feldman pointed me to Bull's new PR campaign. The European HPC hardware manufacturer has started a media campaign that you can read about at and experience for yourself (with videos) at

Some of the campaign, created to look like a guerrilla effort complete with a spray painting Flash web home page, is a little schlocky. Some of it, like the concept videos, has extremely high production value, something we don't often see in ad campaigns in HPC (when was the last time you saw a Cray or an SGI commercial?).

Why do you care? The whole thing (schlock and all) is interesting because it shows the biggest European HPC vendor developing a strategy to bring modern marketing practice from its enterprise client base into a campaign that bridges the enterprise and technical HPC communities.

I believe we'll see more of this as the enterprise side of the HPC market continues to grow and the traditional HPTC vendors (Cray, SGI, LNXI) rush to figure out how to capture more of that market while the enterprise HPC vendors (like HP, Penguin, and Rackable) develop dual-purpose enterprise HPC marketing efforts into HPTC.

At least I hope we do. Then I'll finally be in a career field that's big enough to have its own commercials.

>>NVIDIA's Tesla: 2 TFLOPS on your desk

Earlier this year NVIDIA had several announcements on hardware and its CUDA GPGPU programming framework around its GPGPU strategy as it tried to stave off an assault by the AMD/ATI combo and its Fusion effort.

Now they've followed up with new Tesla GPU cards and integrated systems. The Tesla GPU is an output-less video card with 128 processors. The company puts the cards' performance at about half a TFLOP.

Tesla comes as a PCIe x16 card that consumes two PCIe power connectors and 170 watts at max. You can also get Tesla-powered workstations in 1 and 2 TFLOPS configurations sporting 2 and 4 Tesla cards each. The card will run you $1,499 while the workstations go for $7,500 and $12,000 respectively.

Not exactly loose change, but I remember my first TFLOPS system; it cost $10 million.

The extended version of this article is packed with links; you can find it at

>>SRC's reconfigurable processors come to AMD

SRC Computers announced yesterday ( that they'll be supporting AMD's Torrenza initiative with their processor accelerator solutions. Recall that Torrenza is AMD's platform for letting 3rd party vendors integrate tightly with its processors using HyperTransport (

The move brings SRC's reconfigurable processors, SNAP interconnect, and what the company identifies as its “high level programming environment” to AMD. From the release:

” 'The addition of SRC to the Torrenza initiative is welcomed. Developers will now be able to use ANSI standard languages to develop applications that use both SRC's reconfigurable hardware and the AMD Opteron microprocessor. We see this combination as very powerful in many markets including financial services, oil and gas exploration, and life sciences,' says Randy Allen, Vice President Server and Workstation Division of AMD.”


John West summarizes the headlines in HPC every day at, and writes on leadership and career issues for technology professionals at InfoWorld and on his own blog at You can contact him at