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

10 words and a link

SGI’s new CEO talks about why he did it

HPC virtualization vendor ScaleMP announces SPEC 2X previous x86 best

Intel shows off first Larrabee architecture pic

The HPC powering Wolfram Alpha

Young women dominate nation’s largest pre-college science fair

New Zealand out to replace aging T3E

Appro’s CPU/GPU HyperPower Cluster

Panasas adds SSD to storage solution

Rackable changes stock symbol, finishes name change

SiCortex announces two new “science to science” webcasts

RapidMind seminars for imaging application vendors

NVIDIA OpenCL driver heads in for certification

Presidential Science Advisor testifies on science in new administration

Growing use of GPUs in life sciences

ISC’09 poised to set attendance record

New Energy Star spec for servers

Timothy Prickett Morgan at The Register has coverage of this week’s newly announced Energy Star specification for servers from the US Environmental Protection Agency:

On Friday, the server makers participating in the EPA effort, which will put Energy Star labels on servers that meet energy efficiency and environmental protection standards, received the detailed specs that allow them to certify their machines and receive a rating. The 1.0 spec for the rating is going to cause a lot of grumbling, as any standard in the server racket does. The first pass of the Energy Star 1.0 spec for servers only includes rack and tower servers with no more than four processor sockets and does not cover blade servers (including the blades and their chasses), fault tolerant servers, server appliances, multi-node servers (such as preconfigured clusters), storage equipment, or networking equipment.

Servers up to 4 CPUs are not HPC. Don’t worry though, that’s coming. The group hopes to announce blades and multi-node clusters by the next release on October 15. More in the story.

Fran Berman to speak at Grace Hopper conference

The Anita Borg Institute for Women and Technology (whose director, Telle Whitney, I met at a conference recently; her passion and commitment are infectious) has announced that HPC’s own Fran Berman will be among those to speak at the 9th annual Grace Hopper Celebration of Women in Computing:

The Anita Borg Institute for Women and Technology (ABI) announced today Megan Smith, Vice President, New Business Development, Google, and General Manager, Google.org and Fran Berman, Professor in the UCSD Department of Computer Science and Engineering and the first holder of the High Performance Computing Endowed Chair in the Jacobs School of Engineering at UCSD will be keynote speakers for the 9th annual Grace Hopper Celebration of Women in Computing (GHC), which will be held Sept. 30-Oct. 3 at the J.W. Marriott Starr Pass Resort in Tucson Arizona.

The second annual Plenary Panel of Technology Executives will include Linda Brisnehan, Vice President Military Support Programs, Lockheed Martin Space Systems Company, Nora Denzel, Senior Vice President and General Manager, Payroll, Intuit Corporation, and Werner Vogels, Chief Technology Officer of Amazon.com.

The GHC is an important tool for increasing the retention of young women in the computing sciences:

The 2008 GHC attracted over 1,450 participants from 23 countries and featured more than 300 presenters. GHC provides technical women with visibility, a sense of community, and critical social networks and mentoring relationships that improve female representation in the field. A survey of last year’s participants revealed that a significant percentage reported an increased passion and commitment to their roles as technical women after attending the conference. The survey also found that attendance had a positive impact on their professional advancement.

Many schools, including my alma mater Mississippi State University, provide scholarships to promising women in their computing departments as part of their efforts to improve retention in these programs. For reasons that aren’t entirely understood, a surprising number of women who start computing majors don’t finish them. The GHC has the potential to change that for some women. I encourage you to find out if your school or alma mater has a program to send its computing women to the GHC and, if so, send them a check for $100 earmarked to support those efforts.

NMCAC Has New Mexico Lawmakers Worried

The state of New Mexico’s $11 million supercomputing project, NMCAC, has lawmakers a bit worried. According to the new New Mexico Legislative Finance Committee report, the center hasn’t yet generated the kind of revenue they expected. All told, NMCAC cost the state $13.8 million to roll out but it’s only brought in roughly $300,000 in cash revenue. They’ve generated other revenue in the form of in-kind services, but those services don’t keep the lights lit.

LFC chairman Luciano “Lucky” Varela, D-Santa Fe, said the supercomputer shows promise but needs to “start generating some dollars.” “It’s all nice and pie-in-the-sky, but we’re waiting to see when it is going to become self-sustaining.”

The machine has certainly been busy since its inception back in January of 2008. Three research universities, two national labs and one nonprofit organization have been submitting jobs at no cost since July of 2008. All the bad being said, NMCAC has quite a few supporters within the state. A representative for Governor Bill Richardson had this to say:

“That report is fundamentally flawed and flat-out wrong on most of its major allegations.” He added the supercomputer initiative “has already surpassed expectations and is off to a strong start.” A one-page response provided to the LFC from center officials said their revenues from sources other than the state are at or above projections. They also said there are 28 contracts pending to stabilize costs.

According to the report, however, the big question is the ongoing support and maintenance costs.

“My questions are, how much more money are they going to be asking the state for to maintain it, to upgrade it, to move it…In these very tight, scarce money times, it’s about how can we make it self sustaining.” If the project can’t pay its own way, the LFC report said, officials involved “may want to consider divesting the state of the supercomputer…” Richardson’s science adviser originally asked the Legislature for $42 million over six years for a 200-teraflop supercomputer project that included gateways to the national laboratories, New Mexico State University and the New Mexico Institute of Mining and Technology in Socorro as well as other connections, the LFC report stated.

The state of New Mexico took a huge gamble with NMCAC. This was the first super-scale, state-owned supercomputing resource in recent history. They are, most certainly, braving new ground in operation high performance computing. For more info, read the full article at TMCnet.

—–

John West is part of the team that summarizes the headlines in HPC news every day at insideHPC.com. You can contact him at [email protected].