Visit additional Tabor Communication Publications
November 11, 2010
Next week's Supercomputing Conference (SC10) in New Orleans promises to be a lively affair. With the economy on the mend, the rise of new technologies like GPGPUs and cloud computing, and the prospect of exascale systems just over the horizon, HPC vendors and users will have plenty to talk about at this year's premier supercomputing event. As usual, the challenge for the conference-goer is to figure out the handful of sessions and activities to attend, given the hundreds on the schedule. What follows are some suggestions for an itinerary that tries to hit some of the event highlights.
Learn Me Some HPC
If you're a student of HPC or just want to know more about the technology, and are arriving the weekend before the main event, there are three days chock full of tutorials, workshops, and education sessions on Saturday, Sunday and Monday (Nov 13 - 15) before the exhibition opens. The problem is there are too many to chose from, and since the multi-hour sessions overlap, you can only attend perhaps five or six. Some of the classes are quite specific, such as the tutorials on CUDA, OpenCL, MPI, cluster construction, computational physics and so on. But if you're a newbie to high performance computing, I'd try to find a seat in the Parallel: HPC Overview session on Saturday morning. It's not exactly HPC for Poets, but it does look like a nice overview of what makes supercomputing tick. Also, if you're looking for a broader perspective of supercomputing trends, try and catch the The Past, Present and Future of HPC session on Sunday afternoon, presented by HPC veterans Jack Dongarra and Ian Foster.
The SC keynote address this year will be delivered by Clayton Christensen, who will talk about how to create growth via disruptive innovation. Christensen, who coined the term disruptive technologies in 1995, is a Harvard professor that has developed business models for managing the innovation process. Although not a supercomputing expert, Christensen should have plenty to offer the HPC crowd. If you're a vendor, or a would-be vendor, this is not to be missed.
Bring on the Exascale
Although exascale systems are several years away, forward-leaning HPC types are already looking ahead to the prospect of exaflop supercomputers. There are more than a dozen sessions scheduled on the topic this year, but if you had to pick one, check out the panel discussion on Tuesday afternoon: Exascale Computing Will (Won't) Be Used by Scientists by the End of This Decade. The prospect of hitting an exaflop by 2018 will be debated by an impressive cast of HPC experts including William Gropp, Peter Kogge, Burton Smith, Horst Simon, Bob Lucas, Allan Snavely, and Steve Wallach. Marc Snir will moderate. Should be a lively little discussion.
For those interested in what's new on the cloud computing front, there are a number of sessions to be had (I counted 18). Most, though, are research papers, workshops, and the like, which gives you a pretty good indication of the state of cloud computing in the HPC arena right now. Nevertheless, if you'd like to find out how this fast-moving technology could impact supercomputing, I'd encourage you to stop by the Tuesday research poster: High Performance Cloud Computing: Why Not?!. The research authors will present their work on virtualized HPC and offer some recommendations about optimal use of this technology under different scenarios.
I think we all know SC10 will be swimming in GPGPUs. Following the recent news of Tianhe-1A, the 2.5 petaflop system installed in China, you can be sure this technology is going to get plenty of attention at the conference this year. With GPU computing hitting the mainstream, there are a lot of sessions focused on their use in specific applications and deployment schemes. And the NVIDIA booth, of course, will be hosting a whole series of their own talks and activities around the technology.
For a more critical GPU computing discussion, I'll point you to the Wednesday afternoon panel session: Toward Exascale Computing with Heterogeneous Architectures, which promises to put the GPU acceleration technology in some perspective against alternative heterogeneous platforms. Panel members include reps from Intel, AMD and NVIDIA, as well as Convey's Steve Wallach, so prepare for a brawl.
A List of Lists
The November TOP500 rankings will be announced this week, and although everyone expects the new Tianhe-1A super to capture top honors, there may be a few other surprises as well, GPU-accelerated or otherwise. In addition to the TOP500 rankings is the annual unveiling of the Green500 list, which represents the most energy-efficient supercomputers on the planet. This year will also inaugurate a new list, called the Graph 500, which is an attempt to rank the fastest machines for data-intensive HPC using graph algorithm-based benchmarks. The BoF session on Wednesday evening will reveal the first Graph 500 winners.
Thank God It's Friday Panels
After the vendors have gone home on Thursday, there's still some fun to be had. The Friday panel sessions are usually among the best at SC. This year, as usual, there are four excellent sessions, but because of the time overlap, you can only attend two. My picks would be the Disruptive Technologies for Ubiquitous High Performance Computing panel on DARPA's UHPC program and the Return of HPC Survivor: Outwit, Outlast, Outcompute, which is set up as a sort of American Idol contest for exascale supercomputing designs. A good way to end what will hopefully be a fun-filled week for HPC enthusiasts.
Posted by Michael Feldman - November 11, 2010 @ 3:46 PM, Pacific Standard Time
Michael Feldman is the editor of HPCwire.
No Recent Blog Comments
Contributing commentator, Andrew Jones, offers a break in the news cycle with an assessment of what the national "size matters" contest means for the U.S. and other nations...
Today at the International Supercomputing Conference in Leipzing, Germany, Jack Dongarra presented on a proposed benchmark that could carry a bit more weight than its older Linpack companion. The high performance conjugate gradient (HPCG) concept takes into account new architectures for new applications, while shedding the floating point....
Not content to let the Tianhe-2 announcement ride alone, Intel rolled out a series of announcements around its Knights Corner and Xeon Phi products--all of which are aimed at adding some options and variety for a wider base of potential users across the HPC spectrum. Today at the International Supercomputing Conference, the company's Raj....
Jun 19, 2013 |
Supercomputer architectures have evolved considerably over the last 20 years, particularly in the number of processors that are linked together. One aspect of HPC architecture that hasn't changed is the MPI programming model.
Jun 18, 2013 |
The world's largest supercomputers, like Tianhe-2, are great at traditional, compute-intensive HPC workloads, such as simulating atomic decay or modeling tornados. But data-intensive applications--such as mining big data sets for connections--is a different sort of workload, and runs best on a different sort of computer.
Jun 18, 2013 |
Researchers are finding innovative uses for Gordon, the 285 teraflop supercomputer housed at the San Diego Supercomputer Center (SDSC) that has a unique Flash-based storage system. Since going online, researchers have put the incredibly fast I/O to use on a wide variety of workloads, ranging from chemistry to political science.
Jun 17, 2013 |
The advent of low-power mobile processors and cloud delivery models is changing the economics of computing. But just as an economy car is good at different things than a full size truck, an HPC workload still has certain computing demands that neither the fastest smartphone nor the most elastic cloud cluster can fulfill.
Jun 14, 2013 |
For all the progress we've made in IT over the last 50 years, there's one area of life that has steadfastly eluded the grasp of computers: understanding human language. Now, researchers at the Texas Advanced Computing Center (TACC) are utilizing a Hadoop cluster on its Longhorn supercomputer to move the state of the art of language processing a little bit further.
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?
Join our webinar to learn how IT managers can migrate to a more resilient, flexible and scalable solution that grows with the data center. Mellanox VMS is future-proof, efficient and brings significant CAPEX and OPEX savings. The VMS is available today.