Visit additional Tabor Communication Publications
December 19, 2008
Petaflops supercomputing dominated much of the HPC news in 2008, but the year also witnessed the rise of GPU-accelerated computing and the fall of Linux Networx.
Hit: Parallel Programming Put on the Front Burner.
For the first time, big-name vendors and academic institutions got behind a concerted parallel programming R&D effort with the establishment of two Universal Parallel Computing Research Centers. Intel, Microsoft, UC Berkeley and the University of Illinois at Urbana-Champaign (UIUC) teamed up to figure out how to push multicore/manycore programming into the mainstream. The vendors have chipped in with $20 million, and an additional $8 million will come from UIUC. UC Berkeley has applied for $7 million from the state. Not exactly the Manhattan Project, but a good start.
Miss: Linux Networx Goes Belly Up
In February, the assets of Linux Networx were acquired by one-time rival, SGI. For some extra irony, SGI is now led by former Linux Networx CEO Bo Ewald. The once high-flying HPC cluster maker succumbed to an increasingly competitive marketplace and a customer procurement model that leaves little room for error. RIP LNXI.
Miss: U.S. Science Funding Hits a Political Wall
Partisan politics once again sabotaged U.S. science and technology research funding in 2008. As I wrote in February: "Despite bipartisan consensus to support the President's American Competitiveness Initiative (ACI), which was designed to double federal funding for science education and research at the National Science Foundation (NSF), the National Institute of Standards and Technology (NIST), and the Department of Energy Office (DOE) of Science over the next seven years, the amounts allocated in the FY2008 Omnibus Appropriations bill fell woefully short. The ACI agenda, most of which was turned into law in the 2007 COMPETES Act, was essentially ignored when it came time to dole out the money for 2008." The good news is that the incoming administration and Congress promises to be more science-friendly, although FY09 funding probably won't arrive until March.
Hit: Cray Puts Intel Inside
In April, Cray, the iconic supercomputer maker, and Intel, the iconic chipmaker, exchanged marriage vows to collaborate on research and product lines. The first born of the new relationship -- the CX1 "personal supercomputer" -- was unveiled in September. Considering AMD's eroding competitive position at the high end of the market and Intel's manycore aspirations, it was a logical move for both companies, but especially Cray.
Hit: The Petaflops Era Begins
In June, IBM and Los Alamos National Laboratory installed Roadrunner, the world's first general-purpose petaflop supercomputer, that is, if you can call doing nuclear weapons simulations general-purpose. Just five months later, Cray and Oak Ridge National Lab deployed the second petaflop super. Both systems used AMD Opteron parts, but the IBM machine relied on Cell processors for the vast majority of its FLOPS.
Hit: GPU Computing Builds Its Case
With teraflop-level GPUs equipped with double precision smarts from NVIDIA and AMD hitting the streets this year, graphics chips are on track to become the commodity vector processor for the masses. Not a lot of deployment yet, but GPU-accelerated personal supers, GPU compilers, the continued penetration of CUDA, and the ratification of the OpenCL standard are rapidly filling out the GPGPU ecosystem. Intel, as always, remains skeptical.
Miss: SGI: Same as It Ever Was
For a company that can never seem to make ends meet, SGI made a lot of news in 2008. While the company was busy reconstructing its visual computing business, extending its partnership with NASA, and building the third fastest supercomputer in the world, its balance sheet remained stubbornly in the red. Last week, SGI jettisoned 15 percent of its employees. As I suggested this week, 2009 could be a make or break year for the company.
Hit: Personal Supercomputing Redux
Speaking of personal supers, Cray was just one of many vendors that re-introduced office-friendly HPC in 2008. The main crop of these deskside systems are being accelerated by NVIDIA Tesla GPUs, and came from the likes of Dell, Microway, Penguin Computing, Colfax, BOXX, Lenova, Velocity Micro, ASUS and others. Even Cray's CX1 can now be equipped with Tesla gear. The jury is still out on the long-term prospects for personal supers, but if R&D types can put a few useful teraflops next to their desks, that's a potential game-changer for the industry.
Miss: Quantitative Financial Models Tank
The financial meltdown of 2008 revealed some startling weaknesses in the quant models and how the community was (mis)using them. Bad management is relatively easy to uncover, especially in hindsight, but the extent to which bad algorithms helped bring down the economy may never be known.
Hit: Startups Defy Economic Gravity
Even with the economy in a nose-dive throughout 2008, new HPC companies continued to pop up. Some of the most interesting ones we covered include: Convey Computer (turnkey reconfigurable computing), Nimbus Services (HPC services), Arista Networks (high performance 10 GigE switches), and Darkstrand (service provider for National LambdaRail network). Can't wait to see what 2009 brings.
Posted by Michael Feldman - December 18, 2008 @ 9:00 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 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.
Jun 13, 2013 |
Titan, the Cray XK7 at the Oak Ridge National Lab that debuted last fall as the fastest supercomputer in the world with 17.59 petaflops of sustained computing power, will rely on its previous LINPACK test for the upcoming edition of the Top 500 list.
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.