CSCS Top Right Frontpage
HPCwire

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Language Flags

Visit additional Tabor Communication Publications

Datanami
Digital Manufacturing Report
HPC in the Cloud
Green Computing Report

Tabor Communications
Corporate Video

A Small Company Builds a Big Supercomputer


Atipa Technologies may not be on everyone's radar. It's a small Kansas company, a division of Microtech Computers, which makes everything from $700 desktop computers to Linux clusters for Atipa's HPC systems.

It may end up on the radar soon. Atipa just got a big deal. It's building a $17 million computer for a Department of Energy laboratory. That's not bad for a Linux cluster company that survived a big shakeout in the Linux cluster business five years ago.

The Atipa computer is destined for the Environmental Molecular Sciences Laboratory (ESML) at the Department of Energy's Pacific Northwest National Laboratory (PNNL) in Richland, Washington. It should go live by October. It replaces the Chinook, a 163 teraflops system that cost the lab $21.4 million five years ago. Moore's Law marches on.

"We're very excited," says Dana Chang, vice president of technology at Atipa. "It's a very important sale for us."

The new computer is a 1,440-node system packed into 42 racks, with a theoretical peak processing speed of 3.4 petaflops and 2.7 petabytes of usable storage. That puts it just ahead of the 2.9 petaflops SuperMUC, built by IBM for the Leibniz Supercomputing Centre in Germany – number six on the November 2012 TOP500 list.

ESML has some specialized computing requirements. It is used by hundreds of researchers around the world to employ molecular-level theory for climate studies, genomics, biological analysis, chemical catalysis, and solid-state physics. It's computationally-intense work and requires a lot of storage capability. On the Chinook, climate scientists and quantum chemists were using one core and using up all the memory they could get.

To provide that power, the Atipa computer relies heavily on Intel's Xeon Phi MIC accelerators. Each of its 1,440 compute nodes use x86 processors and the Xeon Phi accelerators, plus 128 GB memory per node. Its 2.7 petabyte storage is a shared parallel filesystem with 60 GB per second read/write.

Atipa was able to create a system to meet the lab's needs at a great price. "All the bids we received were very similar, not much differentiation [in specs,]" says William Shelton, Associate Director of EMSL. "Atipa had the lowest price. Price became the main selling point."

DOE can drive very hard bargains, notes Chris Willard, chief research officer at Intersect360 Research. Although not a major supplier, Atipa has proved to be a survivor. "That PNNL would buy a system from a company like this is not surprising," he says. "PNNL will often go with a smaller company, often because they provide more specialized bids" with tighter margins. Nevertheless, he thinks the price was a good one for the size of the system, and probably allows Atipa to realize a profit.

The computer is also another validation of Intel's Xeon Phi accelerators – and accelerators in general. Intersect360's Willard believes that the fact that a computer of this size is dependent on Xeon Phi takes the interest in accelerators in general up a notch. While there's been a growing interest in accelerators over the past few years, many companies have still just been testing their capabilities. "Accelerators are moving from a technology that people are checking out to one that people are interested in buying in production systems," he says.

Chang agrees. She says that the Xeon Phi was key to providing the necessary price/performance. "It absolutely reduced the cost per FLOPS," she says. "It was the Intel Phi being more affordable while increasing the performance. And it will become more and more affordable in 2013. Customers are now showing a strong interest in getting an Intel Phi card." Atipa has a long-standing relationship with both Intel and AMD as suppliers.

According to Chang and Dan Mantyla, HPC programmer at Atipa, there's an advantage of being a smaller company rather than a giant public corporation with an eye on quarterly profits. "We're such a small company, not tiny but not a big corporation, and we build every cluster on a per-customer basis," says Mantyla. Chang adds that the company's Midwest roots help. "People feel comfortable talking to us because we don't have a big company attitude," she says. "We're from Kansas."

This may not be the last interesting deal from the folks from Kansas. Chang confirms that Atipa has interest from other potential customers as well. "We have a few opportunities we're working on," she says. "We won't know much until the next month or so."

Whether or not those pan out, parent company Microtech computers can already make an interesting boast. It can say its computer selection includes everything from a $700 PC to a $17 million supercomputer.

Related Articles

ARM Muscles In on Intel's Dominance in Datacenters

HPC Programming in the Age of Multicore: One Man's View

Berkeley Lab Contemplates Stepping Stone to Exascale Supercomputer

Genome Analysis Center Selects Convey for Next-Gen Sequencing

SGI Announces Software Support for Xeon Phi

Sponsored Links

Accelerate your science with Seneca
One of the first HPC providers installing a 4X NVIDIA Kepler K-20 cluster. Invites you to a free evaluation on Seneca’s NVIDIA K20 Kepler cluster, pre-loaded with AMBER, NAMD, LAMMPS

Webinar: Programming Heterogeneous X64+GPU Systems Using OpenACC
Join Michael Wolfe as he compares the advantages and costs of using both low-level models and the directive-based OpenACC model for programming accelerated heterogeneous systems. Registration is free.

High-Performance Computing in Action
Businesses that want to be on the cutting edge of their industries are increasingly turning to high-performance computing (HPC) solutions to handle complex compute processes and speed up their rate of innovation. Download this Executive Brief to see how businesses in energy, life sciences and entertainment put HPC solutions to work in their operations.

May 23, 2013

May 22, 2013

May 21, 2013

May 20, 2013

May 17, 2013

May 16, 2013

May 15, 2013

May 14, 2013

May 13, 2013

May 10, 2013


Most Read Features

Most Read Around the Web

Most Read This Just In

Cray CS300-LC

Short Takes

Building Supercomputers with Raspberries

May 22, 2013 | At some point in the not-too-distant future, building powerful, miniature computing systems will be considered a hobby for high schoolers, just as robotics or even Lego-building are today. That could be made possible through recent advancements made with the Raspberry Pi computers.
Read more...

Running Computational Fluid Dynamics in the Cloud

May 16, 2013 | When it comes to cloud, long distances mean unacceptably high latencies. Researchers from the University of Bonn in Germany examined those latency issues of doing CFD modeling in the cloud by utilizing a common CFD and its utilization in HPC instance types including both CPU and GPU cores of Amazon EC2.
Read more...

Computing the Physics of Bubbles

May 15, 2013 | Supercomputers at the Department of Energy’s National Energy Research Scientific Computing Center (NERSC) have worked on important computational problems such as collapse of the atomic state, the optimization of chemical catalysts, and now modeling popping bubbles.
Read more...

Internet2 Awards Program Seeks Innovative Applications

May 10, 2013 | Program provides cash awards up to $10,000 for the best open-source end-user applications deployed on 100G network.
Read more...

Sponsored Whitepapers

Best Practices in Big Data Storage

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.

Progress in Parallel: the Bull Parallel Programming Center

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.

Sponsored Multimedia

SGI DMF ZeroWatt Disk Solution

In this demonstration of SGI DMF ZeroWatt disk solution, Dr. Eng Lim Goh, SGI CTO, discusses a function of SGI DMF software to reduce costs and power consumption in an exascale (Big Data) storage datacenter.

Cray CS300-AC Cluster Supercomputer Air Cooling Technology Video

The Cray CS300-AC cluster supercomputer offers energy efficient, air-cooled design based on modular, industry-standard platforms featuring the latest processor and network technologies and a wide range of datacenter cooling requirements.

SC12 Editorial Feature HPCwire Soundbite sponsored by ISC

HPC Job Bank


Featured Events


  • June 16, 2013 - June 20, 2013
    ISC'13
    Leipzig,
    Germany

  • June 17, 2013 - June 18, 2013
    Forecast 2013
    San Francisco, CA
    United States





HPCwire Events