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.
For those of you who thought Intel was angling for an HPC play with its upcoming Larrabee processor family, think again.
Vendors in the HPC market might fare better in the recession than other IT sectors, but they're not immune to economic gravity.
Nevermind the cores. Just hand over the cache.
The GPGPU contingent of the high performance computing crowd got another big boost on Tuesday with the release of the first version of the OpenCL standard.
The democratization of HPC is unlikely to happen if every company and institution is forced to build and maintain multi-million dollar datacenters to house supercomputers. But there are alternatives.
In our supposedly tech-driven economy, it's common to hear about computer professionals who have lost their jobs and are unable to find new work in their field. Is the IT industry really that much at odds with its own labor market? Surprisingly, yes.
QLogic Corp. has decided to follow its own path with Quad Data Rate (QDR) InfiniBand.
Barcelona, we hardly knew ye. Today AMD launched its 45nm "Shanghai" quad-core Opterons, sending the ill-fated 65nm Barcelona chips into the microprocessor history books.
The petascale era is in full swing. Yesterday, the DOE announced that the Cray XT 'Jaguar' supercomputer at Oak Ridge has been upgraded to 1.64 peak petaflops.
It seems hardly a week passes without some news of HPC being delivered as an on-demand service. That topic includes everything from in-house grids to commercial clouds, but it's the cloud element that's grabbing the attention of the supercomputing crowd.
Suresh Shukla, a leader and advocate of high performance computing at Boeing, passed away on October 15.
As far as technology maturity goes, GPGPU is just a baby. But there's already an effort underway to produce an industry standard for this new programming model: OpenCL.
HPC Vendors Showing Mixed Success in Faultering Economy
Post Date: November 03, 2008 @ 9:00 PM, Pacific Standard Time
Blog: From the Editor
With the financial environment in turmoil, HPC vendors are holding their own... some more than others..
Access to National LambdaRail's high-speed optical fiber network will soon be available for commercial businesses (and just in time for the biggest recession in decades).
Google CEO Eric Schmidt's recent endorsement of presidential hopeful Barack Obama has caused a minor stir in the tech community. While some wonder if execs at high profile companies should even get involved in national politics, the reality is that the tech community overwhelmingly supports Obama over McCain, from the executive suite to the corner cubicle.
Things look grim for the economy right now. And while much of the IT market will proably suffer, HPC may turn out to be a bright spot.
Will Obama Kill Science? When I saw that headline in National Review Online, I thought it might be a good opportunity to read a fresh perspective of the Dems approach to science policy. Boy, was I wrong.
Today, in an announcement that came as a shock to no one in the industry, Applied Micro Devices revealed it would spin off its chip manufacturing business into a separate entity and focus its efforts on microprocessor design.
It's time to vote for the annual HPCwire Readers' Choice awards.
The ongoing financial turmoil in the U.S. continues to flog the stock market and the credit market. Not surprisingly, economists are in disarray, predicting everything from a mild recession to the end of capitalism.
The stock market's recent volatility is giving some publicly traded HPC companies an interesting ride. While tech stocks, in general, have been taking a beating, at least a couple of HPC companies are bucking the trend.
With the release of Windows HPC Server 2008, Microsoft is attempting to make up for its late entry into the high performance computing market.
While hardware accelerators continue to show impressive performance results for supercomputing workloads, Intel is sticking to its CPU guns to deliver HPC to the broader market.
The confluence of the U.S. financial meltdown and this week's High Performance on Wall Street conference in New York might be one of those coincidences that's trying to tell us something.
Prior to yesterday's announcement of the Cray CX1, the SiCortex SC072 was really the only deskside HPC appliance out there. But the two companies have very different ideas of the role of personal supercomputing.
When Intel and Cray became sweethearts back in April, I never imagined the first offspring from that relationship would be a personal supercomputer. But that's what happened.
Over the last seven years, the U.S. Department of Energy's (DOE) Office of Science has changed the direction of DOE research and has become a model for collaborative supercomputing.
IDC offered some encouraging news for the HPC server market for the second quarter of 2008, while admitting it overestimated the numbers for 2007.
What are the top 5 mistakes the marketing department makes when choosing a brand name?
Netezza marries data warehousing with streaming analytics, and it does it in sexy sort of way, geek-wise.
How many times have your heard the word "ecosystem" in reference to the information technology market? Some people aren't comfortable with the terminology, but I think the analogy to the natural environment is near perfect.
Most product design engineers use HPC, in one form or another, as a fundamental tool for development and testing. But some are still on the desktop -- and that might not be a bad thing.
As expected, Intel dominated the IT news cycle this week with its semi-annual developer forum. The company's upcoming Nehalem processor family was the star of the show, but Intel talked about everything from parallel programming to the next-generation Internet.
Since welcoming Intel hardware into the company's product mix 18 months ago, Sun Microsystems has come out with six Xeon processor-based servers. This week the company added two more, including an HPC-only server.
Tell me if you've heard this one before. IBM is planning to deliver one of the fastest supercomputers in the world, to help unravel the mysteries of the universe. Deja vu?
It's not enough that GPUs are doubling their capability every year or so. Performance demand is such that GPU vendors are increasingly turning to multi-GPU configurations.
After a week of media buzz about Intel's upcoming manycore Larrabee processor, I thought I'd try to get a sense of how the competition -- namely AMD and NVIDIA -- is reacting to the news. If Intel is able to deliver the goods with Larrabee, both its rivals have a lot to lose.
In the shadow of Intel's Larrabee unveiling, AMD announced today that it intends to support the new DirectX 11 standard in its stream computing software development kit.
These are interesting times for the microprocessor industry. At the same time the multicore revolution is happening, we're also seeing the rise of data parallel architectures. Yes, vector computing is back, but this time, it's not just for nerds.
The Spanish have hopped on the Cell processor bandwagon. The Barcelona Supercomputing Center (BSC) is planning to build a hybrid multi-petaflop system based on Cell and Power technology.
The IT industry's focus on energy efficiency might seem like a "Mom and Apple Pie" type of pursuit, but there may a darker side to the trend.
The summer months tend to be slow for HPC news, but NVIDIA is helping to liven things up a bit.
IBM's upcoming Power7 chip is headed for multi-petaflop stardom. But energy efficiency might be a real challenge for this processor at the petascale level.
Based on their latest financial reports, AMD and Intel are on very different trajectories. In the latest quarter, AMD lost almost as much money as Intel made. That's bad news for AMD investors, since Intel made a ton of money in the last 3 months. Oh, and along with AMD's dismal earnings report, the company also announced a new CEO.
A lot of industry people in the know are predicting that Moore's Law will come to an end sometime in the next decade. Then what?
Georgia Tech HPC director David Bader calls the Cell Broadband Engine "a processor ahead of its time." That usually means it needs more software.
In HPC, there has always been a tension between general-purpose and special-purpose architectures. That tension reflects two facets of the market: to apply HPC to more application domains and more users, and to increase performance for the most demanding applications. With a sort of schizophrenic behavior, HPC exploits Moore's Law's for all it's worth, and then, unsatisfied, tries to find a way to beat it.
On Tuesday, Intel and DreamWorks announced an alliance to "revolutionize" 3-D animation technology. Although no financial terms of the deal with DreamWorks were disclosed, apparently Intel made them an offer they couldn't refuse.
At a recent Q&A roundtable for journalists, Intel senior VP Pat Gelsinger laid out his vision of the future world of computing. Not surprisingly, Intel was at the center of that world.
As Intel continues to flesh out its multicore processor roadmap, Anwar Ghuloum, principal engineer with the company's Microprocessor Technology Lab, is already encouraging software developers to begin designing applications for manycore processors -- architectures that contain tens, hundreds or thousands of cores.
On Monday, an article in CNews, a Russian IT publication, reported that 100 servers based on the home-grown Elbrus-3M microprocessor would be delivered to its "customers" later this year. The article stated that 0.6 teraflop systems will be built from the Elbrus-3M servers and characterized the new machines as "entry level supercomputers."
Forget curing cancer, solving global warming, or unraveling the origin of the universe. They've finally found the real killer app for supercomputing: advancing chocolate science.
China Ditches Home-Grown Chips in New Supercomputer
Post Date: June 24, 2008 @ 9:00 PM, Pacific Daylight Time
Blog: From the Editor
The largest supercomputer in China, a 160 teraflop Dawning 5000A supercomputer, will be installed at the Shanghai Supercomputer Center (SSC) in November, and will use quad-core Opteron processors.
Petabyte-sized data volumes are forcing researchers to rethink how to perform scientific inquiry. Is it, as Wired magazine says, "The End of Science"?
I'll be heading out to Dresden, Germany in a couple of days to attend the annual International Supercomputing Conference (ISC) and immerse myself in all things petascale.
In all the excitement about the Roadrunner petaflop announcement this week, a bunch of other HPC news got pushed aside. One item that caught my eye was the announcement by the Canadian High Performance Computing Virtual Laboratory (HPCVL) that it had purchased a cluster made up of 78 Sun SPARC Enterprise T5140 servers, which is not a product you hear much about in the HPC space.
This week's achievement of the Linpack petaflop milestone by the IBM Roadrunner was widely predicted, but nonetheless, impressive. Last year at this time, the number one system was Lawrence Livemore's Blue Gene/L at 280 teraflops, and only two other systems -- the Cray XT4/XT3 supercomputer at Oak Ridge and the Cray Red Storm system at Sandia -- made it past 100 teraflops. In fact, the raw computation power of the Roadrunner exceeds the aggregate performance of the top 10 system in June 2007.
The cloud computing meme is permeating practically all areas of computing these days, including HPC. Will the cloud replace the grid as the new paradigm for delivering high performance computing?
With a bigger GDP than the United States, the European Union certainly has the economic wherewithal to field a top tier high performance computing infrastructure. After taking a back seat to the U.S. and Japan in high end scientific computing for the past couple of decades, the Europeans now seem intent on playing in the deep end of the supercomputing pool. The renewed interested is exemplified by the Partnership for Advanced Computing in Europe, whose mission is to build a world-class pan-European high performance computing service.
It was almost exactly one year ago when Cray announced it was lowering its 2007 revenue projections after it learned that AMD would not deliver its quad-core Opteron 'Budapest' processor on schedule. Little did anyone know at the time that the Budapest slip was just a prelude to the larger Opteron fiasco that would play out over the next six months.
With HP's rollout of the new ProLiant BL2x220c G5 today, the company has an answer for IBM's recently announced iDataPlex server. Both are extra-dense server architectures designed for scaled out datacenters. That means these boxes are aimed at cloud computing, Web 2.0 and high performance computing, the current hot markets in IT.
While researching last week's blog on the H-1B topic, I came across an interesting 2007 report from the Urban Institute that challenges the conventional wisdom about the decline of science/engineering education. The report questions the assumption that the U.S technology workforce is inadequate.
The Texas Advanced Computing Center has a great story this week about using supercomputers to perform prostate surgery. My prostate happens to be my third favorite organ and I'm not all that comfortable with a human being fiddling with it -- especially one with a knife in their hand. So this recent development is welcome news to me.
On Tuesday, I wrote about the difficulties that Japan and the UK are having in finding engineers for their local industries. For the flip side of that discussion, today I'm going to talk about how the H-1B visa program continues to agitate the tech community in the US.
The second phase of the Distributed European Infrastructure for Supercomputing Applications (DEISA) initiative was announced earlier today. Like its US-based TeraGrid counterpart, DEISA links up regional supercomputing centers in order to create a common HPC resource for the research community.
If you thought the US had problems finding qualified engineers, look at what's happening in other countries. Even Japan and the UK are reporting that they can't produce enough engineers to fill local demand.
The release of the second beta for Windows HPC Server 2008 was announced on Microsoft's Windows Servers blog site over the weekend. Ryan Waite, Micrsoft's Group Program Manager for HPC says they signed off on the Beta 2 release last Friday.
If you read just one HPCwire article this week, be sure to catch John West's profile of the National Visualization and Analytics Center. The center is developing visual analytic tools for the U.S. Department of Homeland Security. The work is particularly interesting because it fits into the category of "Edge HPC," Tabor Research's term for HPC that lies outside the traditional science and engineering realm.
Earlier today IBM announced the new BladeCenter QS22, a new blade server that incorporates the latest Cell processor, the PowerXCell 8i. While the new name might not exactly roll off your tongue, IBM has managed to address one of the Cell's major technical shortcomings (at least for the HPC crowd), namely much better double precision floating point performance.
The Mathworks has integrated the company's Parallel Computing Toolbox with two MATLAB optimization tool sets: the Optimization Toolbox and the Genetic Algorithm and Direct Search Toolbox.
Late Monday, AMD announced an organizational shakeup, which included the creation of a new centralized engineering organization and the resignation of two top execs.
On Wednesday, AMD presented its revised server processor plans for the next couple of years. The roadmap included the upcoming 45nm Shanghai chip, new six- and twelve-core Opteron processors, and the next-generation socket for DDR3 and PCIe Gen 2. AMD's new path also gives us some idea why Cray decided to play nice with Intel.
A day after SGI said NASA would be installing a 245 teraflop Altix ICE machine at Ames Research Center, the space agency announced it would be teaming with SGI and Intel for their next generation petascale supercomputer, called Pleiades.
Today SGI announced that NASA has selected a 245 teraflop Altix ICE supercomputer for the space agency's next major HPC system. Later in the day, the company posted a $40 million quarterly loss.
Lawrence Berkeley National Laboratory and Tensilica Inc. have announced a partnership to research exascale supercomputing design. The program will combine LBNL's supercomputing smarts with Tensilica's expertise in microprocessor technology.
I got my share of both condemnation and praise from last week's rant about our anti-intellectual culture. I'll save the attaboys for my personal file, but I'd like to share a couple of the more coherent critical responses I received...
HP Labs seems to have come up with something pretty cool. Earlier today, researchers there claimed they'd proven the existence of the "memristor," the fourth fundamental type of electrical circuit.
It's funny how events are always seen as inevitable after they happen. That's the feeling I got from Monday's announcement of the new Cray-Intel alliance. The two companies have joined forces to research and develop multi-petaflop HPC technology for the next decade. It makes sense the iconic x86 chipmaker should hook up with the iconic supercomputer maker, especially considering that manycore computing and the ensuing programming challenges are forcing ...
If you're a regular reader, I'm sure you've noticed that we've done a major upgrade to the HPCwire website. Along with our regular Features section and breaking news (now called Off the Wire), we've included a Top Headline area that aggregates important HPC stories from other publications. We've also added a blog section, where yours truly will continue to add some personal perspective using my 'From the Editor' platform, and ...
"Over 75 percent of Americans don't know they're alive." I half expect to see such a headline someday as yet another example of how poorly educated the U.S. citizenry has become. It's not quite that bad yet, but research has consistently shown us how uneducated students and working adults are in this country. The data reflects not just a lack of education, but a lack of commitment to ...
If you've been listening to the financial news for the past six months, the future seems pretty grim. Intel and IBM seemed relatively unscathed by the all the doom and gloom talk. But AMD is another story.
Get ready for a new and improved HPCwire. We're getting set to launch a completely revamped Web site for the publication, which will include lots of new editorial content, better navigation, RSS feeds, interactive discussions, and a state-of-the-art Job Bank.
Has SiCortex found the right formula for the personal supercomputer? Introduced in November 2007, the company's Catapult SC072 is a deskside mini-cluster that can be plugged into a standard wall outlet. Positioned as the entry-level system in the SiCortex family of MIPS processor-based supers, the Catapult is attracting the attention of some big names in the HPC universe.
HPC vendors seem to have awakened from their winter slumber. A trio of notable products were released into the spring sunshine this week: the first QDR (40 Gbps) InfiniBand adapter from Mellanox; an on-demand HPC development platform from Interactive Supercomputing; and, from newcomer ScaleMP, a flash module that aggregates x86 servers into a virtual SMP.
In a recent report by Forrester Research, analyst Frank Gillett makes the case that HPC and grid computing are not generating broad interest in the enterprise. He comes to the conclusion that vendors should emphasize customer business solutions rather than technology themes. Editor Michael Feldman offers his take on the analysis.
Procter & Gamble's Adventures in High-End Computing
Post Date: March 20, 2008 @ 9:00 PM, Pacific Daylight Time
Blog: From the Editor
Software is one of Tom Lange's favorite subjects -- or least favorite, depending on his mood. Lange heads the modeling and simulation group at Procter & Gamble and is responsible for enlisting computer technology to help develop the company's vast array of consumer products. He spoke last week at the HPC Horizons Summit in Palm Springs, to talk about his company's use of scientific computing technology.
Formula One racing seems to be on an HPC tear lately. Last week, we covered the purchase of an Appro system for Renault's F1 Team. This week, we look at how Red Bull Racing is using Platform LSF to get the most out of their three cluster systems.
There seems to be a general consensus that the datacenter needs to settle on a unified network fabric. The question is, which one? Both Ethernet and InfiniBand vendors have staked claim to unifying the datacenter on their favorite technology.
Save for the occasional article in the mainstream media about how supercomputers have predicted climate changes or discovered some mystery of the universe, most of high performance computing is hidden from public view. The missing element in most stories about supercomputers is how they relate to the human condition at the scale of the individual. But new applications may be on the way that make HPC more personal.
The decline and fall of Linux Networx may serve as a cautionary tale to other struggling HPC vendors. What happened to the feel-good HPC cluster company of 2000-2006?
With AMD fighting to regain profitability in 2008, what will become of the company's efforts to maintain its presence in the lower volume high performance computing market? Editor Michael Feldman talked with David Rich, director of marketing for HPC at AMD, to get a sense of the company's strategy for its high-end computing products over the next couple of years.
In typical forward-thinking California fashion, the folks at Lawrence Berkeley National Laboratory are already looking beyond single petaflop systems. LBNL researchers have started to explore what a multi-petaflop computer architecture might look like, pointing out that power and system costs will constrain how such machines can be built.
For the first time in decades, the majority of the American electorate will have a chance to choose the presidential nominees of both major parties. As we head into the 22-state Super Tuesday presidential primaries on February 5th, this might be a good time to take a look at the candidates' views on science and technology issues.
In response to last week's "Flat Earth" commentary, I received several thoughtful letters. One of the most interesting was from Enda O'Brien, the founder and director of Parallel Programming Services in Ireland, who argued that the world is not nearly flat enough. If it were, he says, salaries of technology workers would be much more globally equitable than they actually are.
Globalization in the 21st century is rapidly leveling the economic playing field and a number of respected analysts believe that science and technology competency will be the criteria that separates the winners from the losers.If so, Americans may be in for a rough ride.
Cloud computing, the scaled-out manifestation of grid computing, is casting a growing shadow on the industry these days. Everyone, it seems, wants in. Is HPC ready to make the jump?
In the relentless drive for more compute power, the new year will see a plethora of new multicore processors, faster interconnects, and bigger machines. But 2008 will be more of a consolidation year for HPC as OEMs and users catch up to the new technology introduced in 2007.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
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?