• NVIDIA Launches Search for Next Great GPU Research Projects, http://bit.ly/2scLzQ
• China Daily reports Intel/AMD hybrid super at 1 PFLOPS theoretical peak, http://bit.ly/3sv4om
• Brocade pushes for 100 GbE in 2010, http://bit.ly/2s8bJo
• Italy joins big dogs list in PRACE partnership, http://bit.ly/1A7XqU
• Oracle braces for EU objection to Sun acquisition, http://bit.ly/2gh9gX

And thanks to everyone who took part in our HPC Community Leadership Award: nearly 1,000 of you! We’ll be announcing the winners during SC in a couple weeks.

Related posts:

• What to read on insideHPC this week
• What to read at insideHPC this week
• What to read at insideHPC this week

Since its 2007 debut, the Green500 has ranked only the energy efficiency of the world’s 500 fastest performing supercomputers. Now, the Green500 is expanding the definition of a supercomputer to include a wider spectrum of the high-end computing world with the “Little Green500″ list and opening its doors to innovation with two new exploratory lists: the”‘Open Green500″ and “HPCC Green500.”

Most of our readers will know that the canonical Green500 list ranks systems on the Top500 list by energy efficiency.

The more-inclusive “Little Green500″ list will broaden the definition of a supercomputer to include any machine used for commodity supercomputing that could have made the TOP500 within the past 18 months.

And the two new lists are aimed at getting people to think more broadly about both architecture and measurement

Two new exploratory lists also are now open for submissions. The Open Green500 will allow supercomputers to use a combination of single-precision and double-precision to generate a correct double-precision result for LINPACK. “We hope that this will stimulate innovation by allowing hardware manufacturers to explore novel architectural solutions while numerical analysts compose novel algorithms to perhaps create a synergistic multi-grid method for LINPACK,” Feng said. Similarly, the second exploratory list will use the High Performance Computing Challenge (HPCC) benchmark for its performance results.

More in the release.

Related posts:

• Green500 Deadline
• IBM dominates the 5th Green500 list
• Energy Efficiency Improvements in HPC

Platform ISF Adaptive Cluster is a new product offering in the Platform ISF family that enables organizations to create an internal cloud for running HPC applications and is integrated with Platform LSF and Platform Symphony. The product dynamically changes the operating system and associated application stack on demand thereby maximizing resource usage and service levels based on workload. It does this by automatically provisioning Linux and Windows operating systems from bare metal or initiating the right guest operating system within a virtual machine container on the compute nodes upon application demand.

To ensure application availability and guaranteed performance, many enterprises employ multiple HPC clusters, each assigned to run a specific application that requires either a Linux or Windows environment. Inevitably this results in cluster silos and “cluster sprawls” with low utilization and high management costs. Platform ISF Adaptive Cluster consolidates the previous application silos into an internal cloud environment that dynamically provisions Linux and Windows operating systems based on application demand to maximize resource usage and service levels. Utilization rates of separate server clusters typically ranging between 40 – 50 percent – can be increased to 80 – 90 percent.

Other providers offer similar capabilities — Adaptive Computing, for example. This is a feature that is not uncommon in enterprise environments, but it is still in limited use in traditional HPC centers.

The other part of this announcement is some plumbing that will allow customers to redirect workloads out to pay-as-you-go resources outside of a user’s cluster during periods of peak demand

Delivered as a solution in the Platform ISF family, the new cloud bursting capability allows Platform HPC customers to access external cloud resources to offload peak workloads from internal HPC infrastructure on a pay-per-use basis. The new solution allows HPC users to use external resources while operating in a managed, private cloud environment as required by enterprise governance and compliance regulations. Platform is working with external cloud service providers to publish a machine image of the Platform HPC management software stack, which includes Platform LSF, Platform Symphony, Platform Cluster Manager and the new Platform MPI, to ensure Platform customers have a single, user-friendly interface to transparently leverage both their internal enterprise infrastructure for HPC and the external cloud for overflow.

Related posts:

• Adaptive Computing puts Moab Adaptive HPC Suite on IBM gear
• Platform revs Symphony, adds support for up to 4k processors
• Platform Computing Releases Platform Symphony 5

To make sure that everyone’s comments are easy to follow, the SC09 conference is encouraging the use of the #sc09 hashtag. As we often do for large events in the community, insideHPC will be including the stream for the #sc09 hashtag right on the website, so you won’t have to go far to learn what everyone at the show is up to.

Related posts:

• Use Twitter or Facebook? Partial to blogs? Now you can get your SC09 fix however you want it
• Twitter experiment: get your insideHPC news on Twitter [UPDATED]
• A quick reference for sustainability events at SC09

The San Diego Supercomputer Center (SDSC) at UC San Diego has been awarded a five-year, $20 million grant from the National Science Foundation (NSF) to build and operate a powerful supercomputer dedicated to solving critical science and societal problems now overwhelmed by the avalanche of data generated by the digital devices of our era.

The new system, named “Gordon” (get it?), will use flash memory for high data throughput and virtualization software from ScaleMP to create large shared-memory “supernodes.” The system is being built by Appro and will be installed in the mid-2011 timeframe. The completed system will have 245 peak TFLOPS, 64 TB of RAM, and 256 TB of flash memory.

“We are clearly excited about the potential for Gordon,” said SDSC Interim Director Michael Norman, who is also the project’s principal investigator. “This HPC system will allow researchers to tackle a growing list of critical ‘data-intensive’ problems. These include the analysis of individual genomes to tailor drugs to specific patients, the development of more accurate models to predict the impact of earthquakes on buildings and other structures, and simulations that offer greater insights into what’s happening to the planet’s climate.”

There is much more in the release, which is quite detailed.

Related posts:

• SDSC experimenting midrange flash-based supercomputer
• SDSC Dedicates New Building
• Today’s flash-based SSDs and their potential in HPC

The CULA team at EM Photonics (the CUDA-accelerated LAPACK solution that we’ve covered before) will be giving away 5 copies of CULA Premium ($395 value) during SC09 from Tuesday, Nov. 17th through Thursday, Nov. 19th.  CULA Premium is loaded with a growing list of CUDA-accelerated LAPACK functions in single, double, single-complex and double-complex precisions (see current list at www.culatools.com/versions/premium) and it also comes with a 1-year ticket support and software updates.

How to get your free copy of CULA Premium: Stop by EM Photonics’ booth #3091 (same hall as the NVIDIA booth) and tell one of the CULA Engineers on duty that you are an insideHPC subscriber!  Only 5 copies will be given away on a first come first serve basis!

(Curious to see how fast CULA is, try the free version at www.culatools.com/versions/basic.)

Related posts:

• LAPACK on CUDA beta available for free, takes refreshing approach to HPC software
• EM Photonics Annoucnces GA Release of CULA Math Lib
• insideHPC readers get a registration discount at NVIDIA’s GPU conference

Purdue University on Monday (Nov. 9) will kick off a national center to develop specialized software to help homeland security and emergency personnel respond to and manage a variety of calamities, from disease outbreaks to economic crises.

…VACCINE is one of two parts of a $30 million parent center concentrating on developing new methods to aid homeland security personnel in preparing for, preventing, detecting, responding to and recovering from terrorist attacks as well as natural and human-caused crises. The parent center – the Command, Control and Interoperability Center of Excellence – is co-led by Purdue and Rutgers University.

Purdue and its team of 15 universities will focus on the visualization sciences, while Rutgers will lead efforts in the data sciences component.

Related posts:

• Purdue named NSF HPC center
• UT Knoxville Wins $16M Math and Biology Center
• NVIDIA names Utah new center of excellence

Yesterday SGI announced financial results for the first fiscal quarter of 2010 (you’ll recall that they’ve shifted their fiscal year). Revenue was up over 100% from last quarter, and as I noted in coverage of those results, is more in line with what one might expect from a union of mostly non-overlapping product lines. The dramatic rise in income is welcomed, but it wasn’t enough to prevent the loss of about $18M in the quarter

Q1 GAAP revenue was $100.1 million, compared to $58.4 million in the prior quarter and $65.1 million in the comparable year ago quarter. Non-GAAP revenue was $122.7 million, compared to $60.5 million in the prior quarter. The company did not report non-GAAP revenue in the comparable year ago quarter. On a sequential basis, GAAP revenue grew by 71% and non-GAAP revenue grew by 103%.

Gross margin was also up to 22.3% from 8.7% in the prior quarter.

“SGI delivered a solid quarter in terms of revenue, gross margin and operations. We increased revenue by 103% and expanded gross margin up to a record 28.4%, both on a non-GAAP basis,” stated Mark J. Barrenechea, president and CEO of SGI. “Operationally, our integration is ahead of schedule in most key areas. We also introduced new products that could expand our addressable market by over $1.7 billion while investing for long-term growth.”

That growth in new products is going to be a significant challenge for the company. They’ve got to tap some of that potential and prove they can produce, service, and sell a product line that is very diverse, and very complex. Sources inside the company have continually complained in recent months that they have no idea when new products are going to be announced, and often get the news first from press releases.

A major announcement coming soon

On the HPC side of their business the company has started hinting at a major announcement to come during SC09. While the company isn’t publicly talking about what the news might be, it is reasonable to expect it will be related to Ultraviolet, the x86-based version of their big ccNUMA shared memory product line. UV is anticipated to offer customers hardware-supported shared memory at large scale with a much lower price premium than the previous generation of Itanium-based shared memory systems.

SGI has been developing this system for years under the leadership of CTO Eng Lim Goh, and the question now is whether they have taken so long to get out  of the gate with a working system that the market has moved past the technology. If it hasn’t, SGI stands to regain some of its preeminence as an innovator in HPC and put some shine back on the star.

Related posts:

• SGI reports Q3 results; posts loss
• Cray reports Q2 results; posts loss
• Sun reports Q3, posts loss

vSMP Foundation for Cloud provides us with unprecedented flexibility in our cloud infrastructure, which is incredibly valuable for the types of projects we are working on that often have very wide-ranging needs,” said Greg Keller, principal of R Systems, Inc. “Using traditional servers or virtualization we are only able to partition our physical systems to accommodate small workloads.  Now, using vSMP Foundation for Cloud, we have the flexibility to dynamically aggregate those same servers to create private on-demand virtual SMP systems for demanding applications.”“The use of virtualization for server aggregation transforms cloud infrastructure into an even more dynamic environment,” said Jie Wu, research manager, technical computing at IDC. “By aggregating memory and CPUs from multiple servers in the cloud on the fly, vSMP Foundation for Cloud provides greater computing capability for users running applications requiring large memory capacity and high memory bandwidth. In addition, the solution’s easy implementation is also a key benefit to the users.”

Key features include:

• On-demand provisioning and re-provisioning of SMP systems for cloud infrastructure leveraging existing provisioning infrastructure
• Seamless integration with systems management and provisioning tools
• Support for any programming model (serial, throughput, multi-threaded or large-memory) without machine boundary
• Aggregation of up to 16 x86 servers to create virtual SMP with up to 128 cores and 4 TB of main memory utilizing Intel’s latest generation of Intel® Xeon® processor 5500 series (codenamed Nehalem)

For more info, read about it at their website here.

Related posts:

• ScaleMP announces vSMP Foundation 2.0
• ScaleMP Announces Qualified Xeon 5500 Server Platforms
• ScaleMP Announces Appro GreenBlade Support

The US software company has refused to offer any concessions to European regulators to meet their concerns about the deal, according to one person close to the process. That has left Brussels close to issuing an official statement of objections, the first step on the path to blocking it, this person added.

…Brussels’ concerns centre on Oracle’s assumption of MySQL, an open source software company that Sun acquired in 2008. Opponents of the deal claim MySQL could become a serious competitor to Oracle’s own core business in the long term.

Sun has basically been on pause in the market since the Oracle bid was announced, losing market share every month that the deal doesn’t close, and laying off thousands of people in an attempt to preserve value for its suitor. From AOL’s Daily Finance

Now Oracle is faced with the issue of whether to walk away from Sun completely. The move would almost certainly involve a legal battle with Sun for breaking a definitive purchase agreement, but Ellison is probably facing a legal fight with the EC if he does not give into requests from its anti-competition authority. In the meantime, the value of Sun assets that Oracles is getting for its $7.4 billion investment is continuing to drop, and at some point the investment will become financially foolish.

An Oracle walkout would be a real mess for Sun, which would have to do major surgery and extensive rebuilding to return to viability without a mate.

Related posts:

• Oracle/Sun deal FAQ: Hardware, MySQL, yes. HPC, Vis, Lustre? Who knows.
• Oracle-Sun deal passes US anti-trust test
• Speculation on the splintering of Sun’s hardware business out of Oracle