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

Mellanox Rolls Out SwitchX FDR InfiniBand with 40 GigE Built In

In this podcast, John Monson from Mellanox describes the company's new SwitchX architecture for high performance computing.

Download the MP3 * Subscribe on iTunes * Subscribe on other Podcast Players.

Maze - Maze Testing and Debugging Tool for Concurrent Applications

In this episode of the Rich Report, I interview Roni Simonian, Founder & Principal of Ariadne LLC. The company's Maze Testing and Debugging Tool for Concurrent Applications is now available for download:

http://kloobok.com

SPM.Python Expoits Parallelism with Fault Tolerance

In this slidecast, Minesh Amin from MBA Sciences presents on the latest release of SPM.Python. Amin was recently awarded several patents for the technology, which now includes support for exploiting parallelism using GPUs by way of PyCuda. With this new feature in place, SPM.Python enables programmers to exploit parallelism in a fault tolerant manner across all three levels of abstraction: servers, cores, and GPUs.

Hadapt's Adaptive Query Execution Speeds Big Data in the Cloud

In this slidecast, I interview Justin Borgman, CEO of Hadapt. This week the company announced an initial round of financing and a set of patents related to Big Data analytics.

“Our patent-pending Adaptive Query Execution™ capability is unlike any other offering in the market,” explains Borgman. “In testing conducted at Yale University, we were able to document query results 50 times faster than Hadoop with Hive and 600 times faster than Hadoop with HBase,” he continues. “Hadapt was born in the cloud and unlike other second generation data warehousing companies, we have optimized our product architecture for big data and big analysis in virtualized environments,” concludes Borgman.

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

Hadapt Closes Initial Financing and Announces Patent-Pending Innovations for Big Data and Big Analytics in the Cloud

Avere's Global Namespace Brings SSD Speed to NAS

In this podcast, Avere President & CEO Ron Bianchini explains how the company's newly announced global namespace (GNS) capability will allow organizations to tame NAS sprawl resulting from enormous growth in unstructured data.

With Avere’s GNS, storage administrators can easily create and manage logical groupings of file-based resources regardless of physical location, presenting clients running NFS or CIFS with simplified and transparent access to data with a single mount point. Moreover, when data is moved from one storage server to another within the network, client access is automatically rerouted with no disruption or downtime. GNS is available immediately as part of the 2.0 release of the Avere operating system.”

For HPC and Cloud users, Avere brings the notion of locality and low-latency to NAS through clustered, hierarchical SSD storage. In my view, it's a scalable approach that's well worth a look.

Download the MP3 * Download the Slides (PDF) * Subscribe on iTunes * Subscribe on other podcast players

Interview: Clustercorp Turns up the Volume on Rocks with New Round of Funding

In this podcast, I interview Clustercorp's newest employees: Mason Katz, CTO, and Greg Bruno, VP of Engineering. As key developers of the Rocks open source software, Bruno and Katz came over to Clustercorp last week to bolster the commercial version of Rocks.

As announced this morning, Clustercorp has just completed its first round of institutional funding, which it will use to scale development and accelerate its successful entry into the cloud management space.

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



In this time-lapse video from SC2003, SDSC builds a live, 128-node Rocks cluster in the Sun booth in approximately one hour and forty minutes.