The Leading Source for Global News and Information Covering the Ecosystem of High Productivity Computing
August 17, 2007
In describing the HPC industry's adoption of clusters, my friend and colleague Chris Willard recently said, "The revolution is over. The revolutionaries won."
That's pretty much right. It's almost to the point where "HPC cluster" is redundant. Most systems -- unless they are very small -- are run as clusters. Major vendors switched over their product emphases, and there has been a proliferation of smaller HPC cluster vendors specializing in Linux for the last 10+ years.
Another compelling trend has been the industry's move away from HPC-specific architectures. With a few notable exceptions (e.g., Cray XT4, SGI Altix, IBM Blue Gene, and arguably various offerings from Linux Networx), the server products used in HPC are the same products used in enterprise computing. They may be repackaged or reconfigured for HPC, but the underlying R&D is leveraged across a broader market.
With these trends in mind, it's caught my attention that three relatively new companies are focusing on tightly-coupled architectures specifically designed for HPC -- Liquid Computing, SiCortex, and Convey.
Liquid Computing is the most established of the three. I first came across them in 2005 as they were ramping up toward the launch of their LiquidIQ architecture. In 2006 they won the HPCwire Editor's Choice Award for Most Significant New HPC Hardware Product or Technology, and they began general availability in October. In March of this year they announced a significant product acceptance with the U.S. Army.
Based in Ottawa, Liquid Computing also seems to be part of a growing Canadian presence among vendors across the HPC ecosystem. Platform (grid and cluster management in Markham, ON), Acceleware (accelerated applications in Calgary), RapidMind (multi-core optimization in Waterloo, ON), and PlateSpin (datacenter virtualization in Toronto) all come to mind immediately. Maybe it's due to superior cooling.
Right behind Liquid Computing is SiCortex, based in Maynard, MA. Under the tagline "TeraFlops from Milliwatts," SiCortex has shipped a few early-access systems, and it looks like they'll be ready to make a move in conjunction with SC07. SiCortex is currently riding the low-power wave, and their architecture does create power consumption advantages. But with their HPC-specific design, I won't be shocked if performance and productivity take over as the main headlines soon.
Convey is still under the radar, and there's not a lot of publicly released information yet. It suffices to say that Bruce Toal has left HP and rounded up some of the old Convex gang. (Presumably Bruce will launch a third company called "Convez" later this lifetime.) And if Convey launches a vanilla cluster, I'll eat an Ethernet cable.
To review: Clusters have taken over, and leveraged architectures rule. Three new companies are bucking that trend. My analysis: Not a bad idea. I still think there are substantial pockets of users out there who are looking for something different. If these companies can compete on delivered productivity instead of raw flops per dollar, one or more could have a significant impact in the market.
-----
Source: Addison Snell, Tabor Research Blog -- http://blog.taborresearch.com/
PSSC Labs PowerWulf Clusters Custom Configured HPC Solutions for Your Needs and Budget
PSSC Labs stands for Professional Service, Super Computers. Our mission bring a superior level of service and support to the HPC community.
FREE Download: "Going Parallel - An Implementation Guide"
Breakthrough performance for MATLAB®, Python and other desktop apps... Get 100X speedups, with less than 10% of the development time. Focus is on enabling familiar desktop tools to virtually execute on parallel servers, clusters, and grids.
The size and diversity of the HPC market in the United States supports a varied set of system providers and integrators. But in Europe, and the United Kingdom in particular, the market has a different shape.
Read More...
PRACE to evaluate petaflops prototypes; Acadamic roundtable discusses the computing industry's talent pool; and WRF benchmark data are released. John West recaps those stories and more in our weekly wrap-up.
Read More...
Since the first patent was issued for a Venetian statue in 1471, 60 million patents have been awarded around the world, with four million patents actively in force today worldwide. And 800,000 new inventions are registered every year. While the data is public, current search tools are inconvenient and inadequate to the needs of professionals. Semantic supercomputing techniques are helping researchers tackle this difficult challenge.
Read More...
Sep 05 | Uppsala University | Swedish researchers are revealing that "intelligent" computer-based methods for classifying patient samples is worthless when it comes to practical problems. Read more...
Sep 03 | Telegraph.co.uk | A new form of three dimensional scans could revolutionise brain surgery within a year, doctors claim. Read more...
Sep 03 | Linux Magazine | In HPC, most attention is paid to utilization and performance, rather than service availability and problem notification. This article focuses on the latter Read more...
Sep 03 | Nature News | What does it take to store bytes by the tens of thousands of trillions? Read more...
Sep 01 | Delaware Online | In the world of supercomputer-powered science, speed is everything, and an open road can lead to the promised land. Read more...
Sep 05 | | The excellent scalability features of Linux, in addition to robust security and performance makes it an excellent choice for server systems, especially in the high performance computing area.
Sep 01 | | The paper outlines the basic steps and tools involved in the process of migrating a desktop application to a parallel environment.
Jun 05 | | As pressure increases on the upstream seismic processing community to deliver ever-higher levels of productivity and efficiency, a new generation of storage solutions will be required that allow the maximum utilisation of high-performance computing (HPC) Linux cluster resources, together with the minimum of management overhead.
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.
Today, HPC organizations are requiring substantially more floating point performance to solve real-world problems. In this podcast, Ben Bennett, ClearSpeed General Manager, discusses how acceleration technology can improve the overall performance of standard x86-based systems...
Get updates and insights on the High Productivity Computing industry delivered driectly to your inbox.