The Leading Source for Global News and Information Covering the Ecosystem of High Productivity Computing
HPC Matters is a joint blog consisting of contributors from the Tabor Communications team on their observations and insights into HPC matters.
June 12, 2008
I recently asked my son, a first year Ph.D. Neuroscience student at a leading New England university, what he knows about high performance computing or supercomputing. His first response was (somewhat kidding), "Do you mean like the old Cray computer that was used in Jurassic Park?"
Well, yeah... but I would have thought that a young researcher might have run across HPC in his post-baccalaureate science courses at Harvard, or during the three years he worked as a senior research technologist in the MassGeneral Institute for Neurodegenerative Disease (MIND) doing mouse brain research, or during his first year of doctoral work. Wouldn't you? Turns out he's never heard anyone on any of his research teams talk about HPC or supercomputers. Nor has the technology's role in the hard sciences been incorporated into the curricula.
I happen to know that his undergraduate institution, the University of Massachusetts, is the same one whose physics department assembled the PS3Gravity Grid, a sixteen PS3 cluster, to explore black holes, and that his current academic institution is home to an advanced computing center with great resources. We won't even discuss Harvard.
Every conference I have attended in the last year has included discussions on the next generation of HPC users, and researchers and vendors alike bemoan the fact that we are not preparing this next generation to use the high performance computing technology all of you rely on. To quote from the announcement for the upcoming International Advanced Research Workshop on High Performance Computing and Grids, June 30-July 4 in Cetraro, Italy: "HPC is viewed as one of the most important technologies of the 21st Century.... Significant amounts of energy and financial resources have been invested by governments and private sectors for building adequate infrastructures and human skills to enable the expected benefits. Yet, despite many spectacular accomplishments of HPC, this tool is still underutilized by many academic schools, industrial companies and businesses."
As the person who's been helping with the tuition for the last eight years, I have a vested interest in raising awareness of the underutilization of HPC in academia, particularly among the undergraduate and graduate student populations. You all have a vested interest as well. So, the question is, what can we, all of us in the HPC community, do to raise awareness and get this knowledge and skill set into the hands of tomorrow's scientists, researchers and engineers?
Posted by Diane Lieberman - June 12 @ 5:00PM
There are 4 discussion items posted.
Dinosaur HPC
Submitted by $user.username on 06/12/2008 - 9:44PM
Jurassic Park used a Thinking Machines computer, not a Cray. Of course, it wasn't a real computer but it had the largest number of lights of any machine produced by TMC. Los Alamos had the largest one of its actual machines.
Post #1
Education and HPC
Submitted by $user.username on 06/13/2008 - 6:03AM
First, academia and industry must deploy accessible and useful machines where students can get to them. Big behemoths need not apply, we're talking thousands of machines here.
Second, deploy them so that they can be used interactively: Big behemoths need not apply, unless they chuck the PBS/LSF/IronGateWithArmedGuard
scheduling discipline.
Finally, make them available outside as well as inside the curriculum. Our best engineers weren't motivated by grades and merit badges, they were turned on by the technology and what they could do with it.
--matt reilly www.bigNcomputing.org
Post #2
Actually, they're both right
Submitted by JohnWest on 06/17/2008 - 2:08AM
In the book the computers used by park researchers were Crays. In the movie Thinking Machines model CM-5 systems are shown (along with Macs and SGIs, including an old Crimson that wasn't plugged in).
Post #3
The missing link
Submitted by icehawk on 06/21/2008 - 4:47AM
I have spoken with several HPC colleagues about the lack of a proper interface between HPC Systems and the scientists that use them. Just because you’re an expert in (fill in the blank) doesn’t mean you can, or even have a willingness, to learn how to port your science into some mathematical algorithm that a computer can understand, or even to use the existing codes that are currently available.
There needs to be a class of people out there that understands the science enough to interface properly with the researcher, while at the same time has a deep enough understanding of HPC systems to properly help port code to that platform. Contrary to popular belief, not all supercomputers are super for all codes. (-:
We don’t need a dedicated HPC systems analyst, nor a dedicated researcher, but instead someone somewhere in between. Someone who won’t go all geek on the researcher and yet at the same time ask questions like, what’s a cache hit?
Imagine a supercomputing center with a staff of dedicated computational experts in Biology, chemistry, physics, etc that are there strictly to interface with the researchers. It would be nice.
Rich
Post #4
And don't forget HP... by JohnWest
Niche CPUs by voqk
kilowatts per hour? by markhahn
multicore programming paradigms by voqk
Re: It's the End of the World as We Know It? by mikeb
The missing link by icehawk
Actually, they're both right by JohnWest
Education and HPC by $user.username
Dinosaur HPC by $user.username
Re: Anticipating the Fall: Application Performance Has Chased Multicore's Speed Right Over a Cliff by tuccillo
Re: Anticipating the Fall: Application Performance Has Chased Multicore's Speed Right Over a Cliff by $user.username
Re: Anticipating the Fall: Application Performance Has Chased Multicore's Speed Right Over a Cliff by $user.username
Innovation and $ by $user.username
Innovation vs innovation by $user.username
Innovation by $user.username
changes in the HPC market by $user.username
While the Microsoft juggernaut has been touting the joys of its new Windows HPC Server 2008, the Linux HPC contingent has been somewhat less vocal of late. But now Red Hat has come up with its version of an integrated cluster solution.
Read More...
Even though the cost of servers still dominates the datacenter budget, storage is actually on a steeper growth curve. HPC storage, in particular, is being singled out as high-growth opportunity. Vendors are scrambling to keep up.
Read More...
Google datacenters most energy efficient; Cluster Resources to demo Moab Hybrid Cluster; Red Hat Linux releases HPC distro. John West recaps those stories and more in our weekly wrap-up.
Read More...
Oct 06 | The Register | Does the HP Oracle Database Machine represent InfiniBand's big chance to break out its HPC niche? Read more...
Oct 06 | BusinessWeek | A body scan can save a lot of time in the fitting room, and fields from medicine to architecture are adopting 3D computing applications. Read more...
Oct 03 | UCSD News | Despite the evolution of computer science over the past 30 years, structural engineering -- hindered by a reluctance to adapt to digital innovations -- has remained relatively unchanged as a discipline. Read more...
Oct 02 | New York Times | Silcon Valley is starting to feel the effects of the credit crunch. Read more...
Oct 01 | Data Center Knowledge | Google today disclosed details of its data center energy usage, confirming that it operates some of the most efficient facilities in the world. Read more...
Sep 04 | | Disk drives are approximately 250 times denser today than a decade ago. This is good news for users who are creating, manipulating and storing more data than ever before. It gives them an opportunity to derive more value from their stored data and lowers the capital acquisition and operating expense associated with that data.
BlueArc's Titan architecture represents an evolutionary step in file servers by creating a hardware-based file system that can scale bandwidth, IOPS, and overall data capacity well beyond conventional software-based devices. With its ability to virtualize a massive storage pool of up to four usable petabytes of tiered storage, Titan can scale with growing data requirements, offering a competitive advantage for businesses, researchers, or other enterprises seeking to better manage data growth while still ensuring optimal performance.