Imagine my delight in discovering that the famous Dead Parrot Sketch has its own Wikipedia entry. God bless Jimmy Wales.

Anyway, on to the news. After a strong showing of initial interest in our job board experiment from potential employers, the job board has dwindled down to just a couple entries. This isn’t something we ever charged for, so I can only conclude that even at free the board isn’t providing value for our readers. And since it takes more than zero time, there’s no sense in doing something with no value.

Soon it will cease to be, where “soon” is defined as when I get around to it, possibly today. My thanks to those who did list jobs, and to those who used it to look for a new job. If you are on either side of the job market, and find yourself in need of a board, I’m told HPCwire still has theirs.

Related posts:

• Announcing the Job Board at insideHPC
• Quick update in the insideHPC Job Board
• Bunch of new HPC jobs on the insideHPC Job Board

Cray is slated to announce its Q4 and FY2009 earnings next Tuesday. Ahead of that announcement it is talking up the success of its XT?m line, the company’s midrange line of supers. You may recall that Cray announced preliminarily neutral financial results early last month.

As momentum builds for its line of midrange supercomputing systems, global supercomputer leader Cray Inc. (NASDAQ: CRAY) today announced that new customers in Japan, Europe and the United States have purchased midrange supercomputing systems from Cray. With the same scalable design and petascale architecture included in the world’s fastest supercomputer, but configured and offered at lower price points, the Cray XT5m and Cray XT6m systems are generating new business wins for Cray in the university, weather, life sciences, and government research and development communities.

Among the early customers for the XT6m are the Institute of Statistical Mathematics, first Japanese customer, and the University of Duisburg-Essen, the first German customer.

The Cray XT6m is the company’s second generation of its midrange supercomputer designed to effectively scale down Cray’s high-end systems while providing the same benefits to an expanded base of users. Upgradeable from a Cray XT5m, the Cray XT6m includes compute blades that feature four compute nodes designed for high scalability in a small footprint and can be configured with up to 96 dual-socket nodes per cabinet. Each compute node is composed of two AMD Opteron(TM) 6100 Series processors (the eight and 12-core “Maranello” platform), each coupled with its own memory and dedicated Cray Seastar2+ interconnect. The compute nodes in the Cray XT6m systems can also be configured with 32 GB or 64 GB DDR3 memory.

Although we won’t know whether this is the case this time, it is not uncommon for company’s to try to get good news out just ahead of unfavorable financial reports. We’ll know next week.

Related posts:

• Cray announces XT6, first customer
• Cray announces “cash for clusters” gimmick, er, buy-back
• Cray completes debt buy back, signals extraordinary financial health

HPC entrepreneur succumbs to apparent heart attack.

Yesterday we had news from our friends that John Mucci had passed away Sunday from an apparent heart attack. John, along with Jud Leonard and Matt Reilly, helped found SiCortex in 2003.

From a profile in Xconomy

Mucci had a long history as a sales executive, and before co-founding SiCortex with Matt Reilly and Jud Leonard around 2003, had been a vice president at Thinking Machines Corporation, where he worked from 1986 to 1994, according to his LinkedIn page. He had also served in several executive positions at Digital Equipment Corp., and had co-founded another startup, TopicalNet (previously known as Continuum Software), according to his bio on the SiCortex site, which is still up although the company announced it was closing its doors last May.

…SiCortex had been funded in part by Polaris Venture Partners. Polaris general partner Bob Metcalfe wrote this in an email tonight: “While working on SiCortex, John and I walked together, he my guide for three years, through the exhibits of the SuperComputing conferences. Everyone there knew and liked John. And I mean everyone.”

Bob is right: I had the same experience every time I interacted with John at SC. Whether he was briefing me on SiCortex products or just talking about the state of HPC today, John was smart, affable, and just fun to be around.

We would like to express our deepest sympathies to John’s family and friends. He will be missed.

Related posts:

• SiCortex IP still on the shelf
• Q&A with SiCortex co-founder Matt Reilly on what happened, and what it means for HPC
• SiCortex Wins Boston Business Journal Green Business Award

IBM today announced major enhancements to its performance management software designed to help CFOs and finance managers accelerate the adoption of business analytics and drive smarter decisions for better business outcomes. The new analytic finance solutions can reduce customer planning cycles in half and reduce reporting cycles from days to minutes.

C-level executives today are looking to Finance to provide more accurate data more quickly, to aid in risk management and forecasting decisions, and to help drive the bottom line. As a result, Finance departments are under increased pressures with higher demands for real-time data production and analysis, increased regulation on reporting and compliance, and reduced budgets. 

To address these issues, IBM has enhanced its performance management solutions

Software investment to drive high-science training for future industry leaders

Paradigm (www.pdgm.com) announced today its Gold Level sponsorship of the 2010 American Association of Petroleum Geologists (AAPG) Imperial Barrel Award Program (IBA). The AAPG IBA program is a highly competitive scholarship program focused on promoting petroleum geoscience training and advancing the careers of geoscience students.

Paradigm will provide participating universities with licenses, support and training for the Paradigm software. This will provide students with the tools they need to complete a technical assessment of field hydrocarbon potential, from seismic interpretation, geological modeling, and petrophysics to reservoir engineering and risk assessment.

"On behalf of the AAPG Imperial Barrel Award Committee

Spectra Logic, celebrating 30 years of data storage innovation, today announced a record-setting quarter for its disk-based backup and deduplication appliances. The positive quarter reflected strong double-digit year over year growth for its enterprise and mid-range data protection solutions, which includes disk, tape, software and services.

Following three consecutive years of profitability, Spectra Logic sustained its profitable track record, posting profits for its fiscal second quarter and first half of 2010, which ended December 31, 2009. The world’s highest capacity enterprise tape library, the Spectra T-Finity was one of many leading products and

Company says new high-end servers will deliver "intelligent performance."

Wireless technology promises energy-efficient chip-to-chip communication.

ScaleMP today announced that Coventry University, an innovative educational institution in the UK, has dramatically reduced costs and improved research capabilities by implementing the vSMP Foundation for SMP software in the Automotive Engineering Applied Research Group. Since deploying the vSMP Foundation software, Coventry University has reduced CAPEX and OPEX, and found the solution easier to manage, offering the flexibility to run different types of applications in its virtualized environments (e.g., OpenMP, MPI, legacy code, etc) as well as significantly better performance than the traditional high performance computing (HPC) system Coventry previously had in place.

To keep up with the increased demand from

Eighty-three percent of IT executives intend to keep their cloud initiatives within their own firewall, according to a survey of delegates by Platform Computing at the Supercomputing Conference (SC'09) in November 2009.  Platform Computing surveyed 95 IT executives across the research, manufacturing, government and education industries to determine why organizations are investing in cloud solutions, their concerns about cloud computing and plans for future initiatives.

Although most organizations are experimenting with cloud computing, the majority (82%) do not foresee cloud bursting initiatives, demonstrating that cloud models are still early in their adoption and that executives are cognizant of the

Largest higher education institution in Brazil uses NetIron MLX routers and wireless LAN solutions as networking foundation

Brocade today announced that The University of São Paulo (USP), the largest public university and research institution in Brazil, is driving advanced communication, business, education, and healthcare applications over its Brocade switching and routing infrastructure. With more than 80,000 students and 6,000 professors on seven distributed campuses, the demand for a performance orientated network is essential to help maintain productivity.

Established in 1934, USP includes 40 educational and research divisions, 43 libraries, five hospitals, five museums, and multiple scientific and cultural centers. The Higher Education Evaluation and Accreditation Council of Taiwan ranks USP in

Free Trial License Available Through May 9, 2010

Zircon Computing has released Version 3.1 of the Zircon Software Product Suite of adaptive middleware and high performance software. The company also announced that it will provide a free trial license for usage in up to 256 cores through May 9, 2010.

“With this release Zircon software delivers several performance improvements, including in-process multi-core optimizations that Zircon’s adaptive load balancer uses to scale the parallel processing of service requests, as well as exposes the powerful zNet service delivery platform to C++ application developers” said Alexander Mintz, CEO of Zircon Computing.

The zNet pattern-oriented middleware delivers a comprehensive array of functionality, including a real-time high-performance computing environment that enables the rapid development of cooperating business

Business Continues to Grow for the Cray XT5m System and the Recently Announced Cray XT6m System

Cray today announced that new customers in Japan, Europe and the United States have purchased midrange supercomputing systems from Cray. With the same scalable design and petascale architecture included in the world’s fastest supercomputer, but configured and offered at lower price points, the Cray XT5m and Cray XT6m systems are generating new business wins for Cray in the university, weather, life sciences, and government research and development communities.

“The Cray XT5m and Cray XT6m systems are industry leaders in terms of compute density and energy efficiency for x86 systems, and our customers – both

• Eighty percent of the Global 100 corporations have chosen Itanium-based servers for their most mission-critical applications.
• More than double the performance of the previous Itanium processor.
• More room for growth with up to an 800 percent increase in system interconnect bandwidth, up to a 500 percent gain in memory bandwidth and up to 700 percent more memory capacity.
• Even better support for mission-critical environments with reliability, availability, and serviceability enhancements throughout the processor and platform.
• Common platform ingredients with Intel Xeon processors foster innovation, design synergy, manufacturing efficiency and flexibility for customers.

Intel Corporation today introduced the Itanium processor 9300 series, previously codenamed "Tukwila," which delivers more than double the performance of its predecessor, boosts scalability and adds

IBM today announced new POWER7 systems designed to manage the most demanding emerging applications, ranging from smart electrical grids to real-time analytics for financial markets.  The new systems incorporate a number of industry-unique technologies for the specialized demands of new applications and services that rely on processing an enormous number of concurrent transactions and data while analyzing that information in real time.

New office would target nation’s fast-accelerating climate information needs

NOAA launches www.climate.gov as portal for climate science and services

            Individuals and decision-makers across widely diverse sectors – from agriculture to energy to transportation – increasingly are asking NOAA for information about climate change in order to make the best choices for their families, communities and businesses. To meet the rising tide of these requests, U.S. Commerce Secretary Gary Locke today announced the intent to create a NOAA Climate Service line office dedicated to bringing together the agency’s strong climate science and service delivery capabilities.

A new kind of Rocky Mountain high.

Leading University Joins Prominent Network of Institutions Focused on Parallel Computing

NVIDIA announced today that it has recognized the University of Maryland as a CUDA Center of Excellence, placing it in an elite grouping of 9 other universities and research organizations worldwide.

The university was selected for its pioneering use of GPU computing and the CUDA programming model across research and teaching efforts within multiple science and engineering departments.

CUDA is NVIDIA’s computing architecture that enables its GPUs to be programmed using industry standard programming languages and APIs, opening up their massive parallel processing power to a broad range of applications beyond graphics.

Other CUDA Centers of Excellence in the U.S. and abroad include Cambridge University, Chinese Academy

Cinch Connectors has announced today that its High Speed Cabling product line is available for purchase through ColfaxDirect.com, a premier reseller of high-performance computing solutions. Cinch's leadership in high performance cabling, combined with Colfax's extensive reach will provide a superior value proposition to the HPC marketplace.

"We feel strongly that the commitment by Colfax to sell the High Speed Cabling product line through their multi-channel network will serve our mutual customers well. Colfax is a highly valued partner in the long term success of Cinch's cable product line, and we look forward to working with them in the future on other critical products." said Mike Salmon, General Manager of Cinch Connectors.

The Cinch high speed cabling

A story in the San Jose Mercury News last week cited officials confident that NASA’s Ames Research Center, located in Silicon Valley, will benefit from the proposed FY11 budget

“We’re highly confident that we’re going to get more money, and we do believe it’s going to enhance employment here, and in Silicon Valley,” said Lewis Braxton III, deputy director of Ames Research Center in Mountain View. He said it is too early to know how many additional local jobs would be created by NASA’s proposed new emphasis on unmanned space missions, climate change research, an extension of the International Space Station and plans to partner with private companies to launch astronauts into orbit.

The potential boon may extend to Ames’s HPC center as well

…Braxton said he could imagine more work for Ames’ supercomputer facility and additional partnerships, such as the center’s cooperation with Google, in which NASA has tapped Google’s expertise to help make its spacecraft more autonomous, and building on partnerships, such as one with Cisco Systems, to integrate the world’s climate change databases.

Related posts:

• NASA Ames Hits 70 Year Mark
• NASA Ames Director Named Best of 2009
• NASA and SGI kick it up a notch: 1 PFLOPS in the Pleiades

Chip maker hopes to bring CPU-GPU processors to servers in two years.

Thierry Manfe has a post at the Sun HPC blogs helping beginners navigate the esoterica of compiler options in the Sun Studio compilers

Sun Studio offers a unique set of optimization features dedicated to processor instruction set that help me squeeze out the best perf out of C, C++ or Fortran code. Yet these options are so numerous that it can be a bit daunting to look into them.

If you are in a rush, you can use the  -fast option. What it really does is triggering a set of other options for maximum runtime performance.

What I like about the post is that it talks about options, and gives those just learning enough information to make their own decisions. To whit

Yet -fast has its own drawbacks. First, the options triggered might change from one compiler release to another. Also, the values for -xarch -xcache -xchip specify the processor for which to optimize, and -fast decides of these values based on the processor on which the compiler runs, which can deffer from the processor on which the code will eventually be executed. This is why I usually stay away from -fast.

Related posts:

• PGI and NVIDIA team up on new CUDA compiler
• 96-Core Multi-Threaded Array Processor Compiler
• NVIDIA’s CUDA Fortran compiler in beta

Last week Dr. Arden Bement, Jr., Director of the National Science Foundation in the US, announced that at the end of his term he will be stepping down to lead Purdue University’s new Global Policy Research Institute. Bement was appointed to a six year term as the 12th director of the NSF by the second President Bush in 2004.

During Bement’s tenure as NSF director, he oversaw the foundation’s annual budget of more than $7 billion that supports the research and education of roughly 200,000 scientists, engineers, educators and students across the United States. As part of the White House’s American Competitiveness Initiative in 2006, he guided initiatives that supported the training of the U.S. work force to operate in a high-tech global economy.

House Research and Science Education Subcommittee Chairman Daniel Lipinski (D-IL) had this to say

“As Chairman of the Research and Science Education Subcommittee, which is responsible for the National Science Foundation, it was my privilege to work closely with Dr. Bement to improve and elevate the NSF. Dr. Bement never forgot either his engineering roots or his time in the Midwest. As an engineer, he could not see a problem without wanting to solve it, and he did a great deal to enhance America’s competitiveness and respond to our STEM education challenges. And as a Midwesterner, he did so with grace and soft-spoken but determined leadership. Our loss will be Purdue’s gain, and I wish Dr. Bement well in his return to academia.”

Related posts:

• House Science Committee budget hearings
• Euro grid takes another step
• HPC R&D Act leaves committee

Main Fourth Quarter Highlights

• Operating loss on a GAAP basis, narrowed to $0.4 million; operating profit on a non-GAAP basis of $0.3 million;
• Net loss on a GAAP basis, narrowed to $0.5 million; net income on a non-GAAP basis, of $0.3 million;
• Cash, cash equivalents and marketable securities as of December 31, 2009 totaled $47.5 million; and
• Introduced 2010 annual revenue guidance range of $66-69

Last week the Transaction Processing Council (TPC) announced that it has defined a methodology for measuring watts per performance in the context of its existing benchmarks

The TPC-Energy specification outlines methodologies for measuring energy consumption in data processing servers, disk systems, and system components associated with typical business information technology environments.

…The TPC-Energy specification augments existing TPC Benchmark Standards, including TPC-C, TPC-E and TPC-H, by outlining requirements to measure and report energy metrics in conjunction with each benchmark. TPC-Energy enables manufacturers to provide power metrics in the form “Watts per performance,” where the performance units are particular to each TPC benchmark. As vendors publish TPC-Energy results, customers will be able to identify systems, via the TPC Web site, that meet their price, performance and energy requirements.

TPC-C is for online transaction processing, TPC-E is for databases, and TPC-H is for data warehouses, so you quickly get the idea these benchmarks don’t directly translate to HPC. Still, understanding the approach may be useful to those looking for ways to extend our benchmarks to include an energy dimension. The TPC-Energy report is available from the council’s Web site.

Related posts:

• TPC video overview of the new energy efficiency spec
• HP and Dell get first servers listed with new Energy Star spec
• New Energy Star spec for servers

Internet2 and NLR have announced that they have signed a memorandum of understanding (MoU) agreeing to collaborate on developing and deploying a unified Cisco TelePresence service offering for the research and education community.  Recognizing the powerful benefits of TelePresence for enabling new levels of collaboration and for breaking down geographic barriers among researchers and educators – nationally and internationally – the two organizations will be combining their respective expertise and reach.

NLR operates a Cisco TelePresence Exchange, located in Kansas City, Missouri.  Over 70 Cisco TelePresence systems in institutions in 12 U.S. states and 3 other countries are already or will soon

Wondering what to read at insideHPC? Some of the most popular posts this week are:

• FY11 President’s budget looks positive for science and tech
• “Anyone walking into a datacenter should break a sweat”
• SPEC MPI2007 benchmark scales to 2048 cores
• NASA goes further with its computational portal project
• BOXX demos hefty SolidWorks Workstations

And if you aren’t subscribed to our email updates already, your friends are probably pointing and laughing at you behind your back. Show them you aren’t hopelessly behind on every day’s technology news by signing up for the full text daily email, or the daily digest with the fancy graphics.

Related posts:

• BOXX Demos Hefty SolidWorks Workstations
• What to read on insideHPC this week
• What to read at insideHPC this week

Just in over the email transom, news of a data center haiku contest being run at ServiceCatalog101.com:

It’s time to boot up the cerebral right side and balance those cortical hemispheres!

For global fame, peer recognition and a beautiful coffee mug featuring YOUR winning Haiku, simply submit your best haiku and win!

We’ll keep the rules super simple:

• No more than three lines.
• Make at least one seasonal reference.
• Keep it data center relevant.

Reference at wikipedia in case you need to brush up on what a haiku is. Here’s an example to get you started

Be strong, blade server
Stay in your rack and burn hot
You are beautiful

You can tweet, email, or post your entry, which must be submitted by Feb 15. Instructions in the post. Winner is announced in the last week of February and gets his or her haiku printed on a mug. Awesome!

Related posts:

• HPC Driving Data Center Architecture Change
• HP Intros 20-Foot Container Data Center
• Data Center World Day 1 links roundup

Addison and Michael talk about how NASA and the NSF are working to bring science codes to cloud computing.