Wednesday, May 25, 2011

The Secret Sauce in Cray’s XK6 Hybrid Supercomputer

In this podcast, Cray's Barry Bolding describes the secret sauce in the company's new Cray XK6 hybrid supercomputer. Scalable to 50 Petaflops, the Cray XK6 is expected to start shipping later this year.

Capable of tens of millions of MPI messages per second, the Gemini ASIC complements current and future massively multicore and many-core processors. Each hybrid compute node is interfaced to the Gemini interconnect through HyperTransport 3.0 technology. This direct connect architecture bypasses the PCI bottlenecks inherent in commodity networks and provides a peak of over 20 GB/s of injection bandwidth per node. The Gemini router’s connectionless protocol scales from hundreds to hundreds of thousands of cores without the increase in buffer memory required in the point-to-point connection method of commodity interconnects.


Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players

Monday, May 23, 2011

IO Turbine Launches Accelio SSD Accelerator for Virtualization



In this slidecast, Rich Brueckner from insideHPC interviews IO Turbine CEO Rich Boberg. The company's newest product, Accelio uses SSD/Flash on the compute server to solve I/O bottleneck problems in VMware environments. When installed on VMware servers, Accelio transparently identifies the highest priority data and offloads IOPS from primary storage to Flash, delivering performance directly to designated virtual machines.

Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players

Wednesday, May 18, 2011

Texas Memory Systems Product Update



In this slidecast, Jamon Bowen from Texas Memory Systems presents an update on the company's SSD storage products. The RamSan-70 “Gorilla” is a high-performance, half-length PCIe card with 900 GB of usable SLC Flash capacity. Delivering 330K IOPS and 2 GB/s of bandwidth, the Gorilla reportedly has higher performance and greater capacity than any other half-length SLC Flash PCIe card.

Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players

Monday, May 16, 2011

Nvidia Tesla 2090, World's Fastest HPC Processor



In this this slidecast, Nvidia's Roy Kim and Andy Walsh provide an update on the all-new Tesla 2090 HPC processor.

In the latest version of AMBER 11, one of the most widely used applications used by scientific researchers for simulating behaviors of biomolecules, four Tesla M2090 GPUs delivered record performance of 69 nanoseconds of simulation per day. The fastest AMBER performance recorded on a CPU-based supercomputer was 46 ns/day. “This is the fastest result ever reported. With Tesla M2090 GPUs, AMBER users from a university department can now accelerate their scientific work as if they had a supercomputer in their own lab,” said Ross Walker, assistant research professor at the San Diego Computer Center, and principle contributor to the AMBER code.


The Tesla M2090 GPU will be available in servers including the new HP ProLiant SL390 G7, which incorporates up to eight Tesla M2090 GPUs in a half-width 4U chassis. Read the Full Story.

Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players

Sunday, May 15, 2011

Cycle Cloud HPC Clusters - 10,000 Cores in 45 Minutes



In this slidecast, Cycle Computing CEO Jason Stowe describes how the company provisioned a 10,000-core, top 114-equivalent supercomputer utilizing its CycleCloud service.
Since 2005, Cycle has helped clients maximize the world's compute resources through its reliable, secure and elastic high performance computing (HPC) solutions, both internally and in the cloud. CycleCloud massively scaled up client resources to perform hundreds of thousands of computational tests in a matter of eight hours. Once the results were produced, the customer could just "turn off" those resources with no further charges. Additionally, the 10,000-core cluster was run on a cost of only $1,060/hour. Leveraging CycleCloud and Cycle's HPC proficiency delivered these stats:

  • Infrastructure: 1250 instances with 8-core / 7-GB RAM

  • Cluster Size: 10,000 cores, 8.75 TB RAM, 2 PB of disk space total

  • Scale: Comparable to #114 of Top 500 Supercomputer list

  • Security: Engineered with HTTPS & 128/256-bit AES encryption



Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players

Monday, May 9, 2011

Samplify Prism FP Takes Floating-Point Compression to the Max

In this podcast, Samplify VP Allan Evans describes the company's advanced floating-point compression technology called Prism FP. Learn more at Samplify.com.

Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players

Sunday, May 8, 2011

DoD Mod Chooses Panasas for Multiple Datacenters

In this podcast, Ron Mowery and Mike Moritzkat from Panasas discuss the company's recent win at the DoD Modernization Office. As part of a subcontract by Lockheed Martin to supply department-wide storage and parallel file system infrastructure to the High Performance Computing Modernization Program (HPCMP), Panasas storage will be deployed at four DoD HPC centers across the country.

Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players

Tuesday, May 3, 2011

ISC11 Launches HPC in Asia Workshop, Spotlight on Exascale



In this slidecast, Rich Brueckner (insideHPC) and Mark Bernhardt (The Exascale Report) interview Martin Meuer, Director of Operations for ISC. This year, the International Supercomputing Conference is launching and all-new HPC in Asia Workshop on the first day of the conference. Meuer describes the goals for the workshop along with plans for an exciting set of Exascale sessions at ISC.

Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players