Archives for posts with tag: Parallel computing

I thought I would point you to some exciting work students are participating in over at Texas Advanced Computing Center.  TACC is mentoring several University of Texas at Austin students in the SC10 Student Cluster Competition to be held this November in New Orleans.  This is great work that the students are doing, and it is great to see involvement from TACC and support from the HPC community.  Here are some brief details of the SC10 challenge to build a system within the power constraints equivalent to only three coffee-makers!

The Student Cluster Competition (SCC) showcases the computational impact of clusters and open source software in the hands of motivated and sleep-deprived students under both a time and power constraint. Uh, no pressure…

During SC10, teams consisting of six students… will compete in real-time on the exhibit floor to run a workload of real-world applications on clusters of their own design while never exceeding the dictated power limit.

Prior to the competition, teams work with their advisor and vendor partners to design and build a cutting-edge commercially available small cluster constrained by the 26 amps available during the conference. Teams must also learn the four open source competition applications and are encouraged to enlist the help of domain specialists.”

It is exciting to see students motivated to work on some of the challenges we face in the industry.  I first met the two student leaders of this six-person group at SC09 in Portland last year. Read the rest of this entry »

Advertisements

There was a recent article in IEEE Spectrum entitled The Trouble With Multicore that gives a really nice background on how multicore processors have taken hold in the computing industry.  I wrote about this article on my blog Multicore in HPC – Where will we stand in 10 years? In doing so, I offered some additional insight into the future of multicore in HPC.

Additionally, and more importantly, I posed the same question to the High Performance Computing group on LinkedIn and there has been lots of great insight posted there as well. So much so, that I thought it would be very beneficial to share some very keen insight others in the industry have shared.

Head on over to my blog at HPCatDell.com, to read some of the highlights of the great discussion going on at LinkedIn.

David Patterson over at IEEE Spectrum has written an article entitled “The Trouble With Multicore.” Kudos to David for a very thorough and well thought out article.  He gives plenty of background on how we arrived at multicore processors, and some of the techniques and challenges that come with parallel processing.  The advent of multicore processing was pretty much a gamble on the part of the semiconductor manufacturers, although their hand was forced due to the power wall associated with increased processor speeds:

“[In 2005] the semiconductor industry threw the equivalent of a Hail Mary pass when it switched from making microprocessors run faster to putting more of them on a chip—doing so without any clear notion of how such devices would in general be programmed. The hope is that someone will be able to figure out how to do that, but at the moment, the ball is still in the air.”

Achieving sustained parallel performance with application codes is a major effort.  In research and engineering communities, we have had increased success, but with that comes a major outlay of time and resources.  Here are a couple of choice quotes in the article that indicate the increased effort it takes to exploit multicore processors: Read the rest of this entry »

A recent article in IT Business Edge asks “Do Processors Really Matter Anymore?”  There is a statement in the article that got me thinking:

It would seem …  that the only users still focusing on clock speeds and overall processing capability are in the HPC market…


To this statement, I tend to disagree.  In HPC there is a focus on processors, but not so much on flops per core anymore.  There is a real concentration on how to leverage parallel computational resources in order to get your application to run efficiently.

Take a look at my comments over at HPCatDell.com.

What do you think?

%d bloggers like this: