Force10 Nabs ExaScale Moniker, Cites Power Efficiency
Exascale computing may be far in our future, but the branding folks at Force10 Networks have it for you on a label today. Announced back in March, 2009, the Force10 ExaScale E-Series of core routers and switches boasts impressive performance specifications, now spiced with leading power efficiency:
Data center power and cooling costs are forecasted to rise significantly in the next several years. Ironically, positive device attributes, such as performance and port density, are driving these higher costs. Consequently, we anticipate data center managers will more closely examine how switches are architected to minimize energy consumption,” said Kevin Tolly, founder, The Tolly Group. “Our testing indicated that Force10 clearly recognizes this ongoing concern and has demonstrated a critically
…
SDSC to Host Data-Intensive Discovery Conference
As part of it’s preparations to deploy the data-intensive Gordon supercomputer in mid-2011, SDSC will host the Grand Challenges in Data-Intensive Discovery conference on October 26-28, 2010 at the UC San Diego campus.
We believe this conference will greatly benefit those doing research in data-intensive fields,” said Michael Norman, interim director of SDSC. “Gordon, and future systems like it, will open new opportunities across numerous areas of research, and it is important that such resources are utilized to their utmost capabilities.”
The conference already has an impressive lineup of speakers and registration is now open. Contributed talk and poster abstract submission are due September 16, 2010.
Video: Preview of SC10 with Sandia’s Barry Hess
In this video, SC10 General Chair Barry Hess gives us preview of the world’s largest gathering of HPC professionals, coming up this November in New Orleans. To see more videos like this, subscribe to the new SC10 YouTube channel.
China’s Petaflop Super Comes Up to Speed
As reported in the People’s Daily, China’s first petaflop (peak) supercomputer, Tianhe-1, is now fully installed and is running user applications 24×7. With 71,680 cores, Tianhe-1 (meaning River in Sky) is ranked at Number 7 on the TOP500 and is installed at the National Super Computer Center in Tianjin. Each node of Tianhe-1 consists of two AMD GPUs attached to two Intel Xeon processors.
The other Chinese systems in the TOP10, Nebulae, is located in Shenzhen and achieved 1.271 petaflops on Linpack, which puts it in the No. 2 spot on the TOP500 behind Jaguar. With it’s NVIDIA GPU accelerators, Nebulae reports an impressive theoretical peak of almost 3 petaflops – the highest ever on the TOP500. Together, the performance of Nebulae and …
Webcast: HPC Advisory Council Accepting Proposals for University Awards
Dan Olds over at El Reg has posted a new webcast interview with HPC Advisory Council Chairman Gilad Shainer. Is the cloud model a good fit for HPC? Their answers answer might surprise you.
Dan’s sense of humor is always refreshing and the HPC Advisory Council is doing great things these days. As part of the discussion, Dan and Gilad also discuss the Council’s new University Award Program, which is now accepting proposals for advanced research around HPC. Prizes include access to compute resources and a case of Red Bull, so check it out.
AMD Demos Fusion APUs in Berlin
In the HPC world, we tend to pay attention to the hottest server chips as they hit the market. But as I’m learning from the Exascale Report, much lower chip power requirements are needed for the industry to reach the next computing milestone of sustained Exaflops.
This week AMD’s John Taylor blogged on some very interesting low-power Fusion chips that the company demonstrated at the IFA computer electronics show in Berlin. Drawing as little as 9 watts, this Accelerated Processing Unit (APU) combines low-power “Bobcat” x86 cores with AMD Radeon discrete GPUs.
Taylor cites expected availability for this APU platform to be Q1 2011. Read the full post here.
Donate Your CPU Cycles to IBM’s Clean Water Grid
We don’t hear much SETI@home or Folding@home these days, but a quick visit to their sites confirms that they are still accepting your donations of compute cycles.
While the notion of Volunteer Computing is not new, this week Big Blue announced Computing for Clean Water, part of a series of IBM Grid Computing projects. It’s a worthy cause if there ever was one, as clean water is a scarce resource for at least 1.2 billion people worldwide.
Like SETI@home and others, the Computing for Clean Water project uses BOINC open-source software for volunteer computing. BOINC is sponsored by the the good folks at the National Science Foundation.
To contribute to the Computing for Clean Water project, sign up …
Whamcloud Grapples for Lustre Mindshare
HPC startups aren’t for the feint of heart. You’ve got to spawn the idea, get financing, and develop the technology while the clock is running. So far, so good for Brent Gorda, CEO of Whamcloud, a new company aiming to close it’s first deal and take Lustre to the next level of performance.
Now comes the hard part — gaining mindshare. As reported in Linux Insider, Gorda is finding that whenever he talks to potential customers about Whamcloud and Lustre, each listener comes away hearing something different:
So I’m finding that I have to spend a lot of my time just talking to people and trying to show them that we are taking our time and trying to do this in a
…
Mercury Marine Rocks the Boat with Windows HPC
If you’re into boating, you know all about Mercury Marine, makers of inboard and outboard motors. As described in this new case study, the company relies on SIMULIA Abaqus software for structural and fatigue analysis of outboard and sterndrive marine engines.
When company executives asked analysts to accelerate an upcoming project, they migrated from a loosely coupled cluster of Linux systems to an in InfiniBand cluster of IBM servers running Windows HPC Server 2008.
We’re able to achieve more realism in our simulations and more refinement in our designs,” says Arden Anderson, CAE Engineer for Mercury Marine. “As a result, we’re increasing durability, reliability, and weight optimization in our products.”
So far, the migration from Linux seems to be going well …
Mainstreaming HPC Hampered by Skills Gap
El Reg has a new feature story on how HPC adoption in mainstream enterprise IT has been hampered by a skills gap.
From our research, it is clear that the impact of HPC on mainstream IT is less to do with technology, and more to do with skills and operations management. Commodity hardware is widely regarded as suitable for HPC, with custom hardware reserved for the most demanding tasks. One of the reasons for this, apart from the direct cost advantage, is the pool of available skills. This has undoubtedly broadened the use of HPC, but the lack of high-end HPC skills has the potential to be a barrier to translating HPC experience into more general IT performance improvement.
Based on research published by Freeform Dynamics, the …
