The Sun HPC Watercooler posted yet another helpful video presentation focused on the trials and tribulations of HPC.  This video is actually the first ‘module’ in a series entitled “An Introduction to Parallel Programming.’  The series will focus on the basics of parallel programming, debugging and general application development tips.

In order to help developers and engineers meet the challenges posed by parallel programming, Sun Microsystems is offering a series of seminars called “An Introduction to Parallel Programming” discussing parallel programming as a fundamental of application development. In this episode, Sun’s Ruud van der Pas kicks things off with a presentation on performance tuning.

You can watch the video here.

Just a quick note to update you on the Green HPC podcast series, which has gotten a tremendous response (thanks!). I’ve added a transcript of the first episode, in case you’re more of a reading person than a listening person. If you don’t know about the series yet, take a listen to the first episode.

ANSYS announced earlier this month (though it just showed up on their website, which is vexing) that Alinghi, the Swiss sailing team that has twice run the America’s Cup, is going to be using their software to support vessel design and analysis.

Applying ANSYS software to structural simulation, Alinghi’s goal is to build the lightest, strongest boat possible. The team will use ANSYS fluid dynamics simulation primarily on the hulls, appendages, sails and masts. “With the sheer size and speed of the class of yacht we’re designing, the loads have become enormous. We need to have confidence in our simulations, because if something breaks people may be hurt. We also need our shapes to be optimal for fluid drag and lift purposes, so while the structural engineers want everything to be shaped like an I-beam for optimal structural properties, the fluids people want everything to be streamlined and skinny. Somewhere in the middle is the optimum, and this is where the ANSYS software is so useful. Doing full simulation using the same family of codes within the ANSYS Workbench environment really makes our life easier,” Bungener added.

NCSA has published a video tour of the facility that will be home to Blue Waters (the future IBM petascale system), hosted by Senior Associate Director John Melchi. The facility is mammoth (bigger than a football field, with 30,000 sq ft of 6 foot high raised floor). Check it out here.

In an upcoming episode of the Green HPC podcast series I interview Bill Kramer about Blue Waters and the steps that NCSA is taking to be proactive in manging — and reducing — energy use by that machine.

UK HPC provider OCF (about whom I’ve written before; for example, this article at HPCwire) has announced two new members to its board of directors

OCF plc, the UK’s premier HPC system integrator, demonstrates its ongoing commitment to providing customers with deep HPC technical skills, expertise and knowledge with the appointments of long-term employees Barry Evans, Technical Manager and Russell Slack, Engineering Manager to its Board of Directors.

…Working for OCF since 2002, Barry is Technical Manager responsible for pre-sales, solution architecture and technical strategy recommendations for OCF’s clients.  With an extensive background in systems engineering and systems management, Barry previously held positions with West Corporation, the U.S. Department of Defense and Utell International in the United States.

Russell is Engineering Manager working for OCF plc since 2002 and with its predecessor firm since December 1995.  Responsible for a team of six engineers; Russell has worked his way up from Trainee Engineer, UNIX System Engineer and Senior Engineer with OCF plc.  Russell oversees every part of customers’ post-sales installations which includes: coordinating delivery of customers’ orders to OCF for pre-installation quality testing purposes; investigating the most effective way of implementing solutions; coordinating the delivery and managing the installation of solutions at customers’ sites and the creation and agreement of ‘Statements of Work’ to ensure that the scope of work is clearly understood by all parties.

PRACE announced late last week that they are hosting a seminar in September to serve as a get acquainted session between industry and the coming pan-European petascale computers

PRACE, the Partnership for Advanced Computing in Europe, a consortium of the 16 leading supercomputing centers in Europe, will arrange a seminar for potential industry users in Toulouse, France, on 7.-8. September 2009. The seminar will focus on industrial needs and expectations of the future European High Performance Computing (HPC) Infrastructure.

Close to 150 invited participants, especially decision makers are expected to attend the event including representatives from aerospace, car manufacturing, energy, chemicals, drug design and medical industry, metal industry, bioinformatics, telecommunication companies, finance, insurance, as well as HPC vendors, scientific software developers and service companies, and EC representatives.
During the seminar participants will receive first hand information about the planned PRACE research infrastructure, services and usage.

More info at the seminar web site.

TechRadar has an article featuring ten “interesting and unexpected” laboratory projects IBM seems to be working on.  Many of these projects fall in the unexpected category simply because the vast majority of technology consumers are outside of the realm of scientific computing.  Oddly enough, the first project on their list involves doing real-time stock analysis using streaming data techniques and a rather hefty IBM BlueGene.  IBM is working with TD Bank in Toronto to analyse financial data as it comes across the wire with very sophisticated algorithms.  Its one thiing to perform options pricing based on current market conditions.  Real time analysis [eg, heavily time sensitive... seconds, not minutes or hours] of data can become horribly difficult.

We are researching the ability to process data from a huge number of sources, so this project is using stream computing and the Blue Gene supercomputer,” says Morris. “We are able to get a 21x speed-up than any other system. The messages and sensors in the financial systems are increasing at more than 50 per cent each year – faster than doubling every two years – so something new is needed to analyse this data.”

“The goal of any automated trading system is to reduce the time between the receipt of market data messages and the decision, achieving a very low latency while processing extreme amounts of data,” adds Nagui Halim, chief scientist of the Stream Computing Project at IBM.

This is horribly interesting stuff.  There are nine other projects listed in the article.  They range from Hudson River data, electric cars and cocoa [yup, chocolate].  Read the full article here.

ScientificComputing.com has an interesting article featuring Dr. Phillip Dickens of the University of Maine.  He and his team are diligently working towards building a small compute grid within the state of Maine.  More specifically, he and his team are focusing on climate change. They’re building a grid portal aimed at giving researchers and students easier access to the vital data and tools they need to perform and evaluate climate data.

Our team is developing Maine’s first scientific grid portal that will execute climate-change models and provide high-resolution visualizations of output data in real time for use by researchers as well as students and educators in the state’s public school system. [Phillip M. Dickens, Ph.D.]

They’ve already planned to expand the scientific capability of their grid by working on a partnership with the Jackson Laboratory, a leading genetics research center.

Researchers at Jackson Laboratory will be able to upload data to University of Maine systems for processing, and the supercomputers will then compute the models and send back the visual renderings of the lab’s data — a true collaboration in solving large problems and a first for the state of Maine. This technology makes it possible to solve problems that are too large to execute at either facility alone by distributing the task across both the University of Maine system and the Jackson Lab cluster so that they can execute the task concurrently.

Dr. Dickens and the University of Maine have great plans and high aspirations for their newly formed grid project.  To read more about it, check out the full article here.

Sun, via their BestPerf blog site, has released a series of NAMD benchmarks from their 6048/X6275 series blade servers.  The benchmarks feature 12 blade servers running the popular parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems.  They wanted to especially note the performance over the IBM BlueGene/L.  From the post:

.: The cluster of 12 Sun Blade X6275 server modules was 4.7x faster than 256 processor configuration of the IBM BlueGene/L

.: The cluster of 12 Sun Blade X6275 server modules exhibited excellent scalability for NAMD molecular dynamics simulation, up to 10.4x speedup for 12 blades relative to 1 blade

.: For largest molecule considered, the cluster of 12 Sun Blade X6275 server modules achieved a throughput of 0.094 seconds per simulation step

The systems consisted of two Intel Nehalem’s of the 2.93GHz flavor nad 24GB of memory each.  The software config consisted of SUSE Linux Enterprise Server 10 SP2 [kernel 2.6.16.60-0.31_lustre.1.8.0.1-sm], Scali MPI 5.6.6 and GNU compilers [ gcc 4.1.2 (1/15/2007), gfortran 4.1.2 (1/15/2007)].  They go into a fair amount of detail on which benchmarks were run and how.

See the full writeup here.