Archives for March 2012

Resilience: Another Big Obstacle to Exascale Computing

My Exascale panel discussion at the Structure: Big Data conference has already hit the news. Matthew Ingram from GigaOm has written up a summary of one of the key topics of the discussion: Resiliency in a system with a million nodes, hundreds of millions of cores, and billions of threads. Speaking at GigaOM’s Structure:Data conference, Los Alamos […]

SC12 Newsletter: Housing Site Opens Today!

Clipped from: sc12.supercomputing.org (share this clip)   The March edition of the SC12 Newsletter is out with these headlines: Technical Papers Abstracts due April 20; Full Papers due April 27 State of the Practice (SOP) now part of the technical papers program Tutorial Proposals due April 27 Calling Network Innovators:  Submit to the SCinet Research […]

Video: Application & Optimization of a GPU Cluster in Seismic Data Processing

httpv://www.youtube.com/watch?v=QilAN5jFhXI In this video, Xiangyang Zhang from the China National Petroleum Corporation presents: The Application and Optimization of the GPU Cluster in the Field of Seismic Data Processing. Download the slides (PDF). Recorded at the HPC Advisory Council Switzerland Workshop on March 15, 2012.

First PGI Compilers with Support for OpenACC Now Available

The Portland Group has beta-released its first Fortran and C compilers with support for OpenACC. PGI continues to invest in technologies to make GPU programming more straight-forward and productive for domain experts in science and engineering,” said Douglas Miles, Director of The Portland Group. “The OpenACC standard was developed in direct response to HPC community’s […]

Video: rCUDA, An Approach to Provide Remote Access to GPU Computational Power

httpv://www.youtube.com/watch?v=6PCK58aS85M In this video, Professor Federico Silla and Antonio J. Peña from the Technical University of Valencia Spain present: rCUDA, An Approach to Provide Remote Access to GPU Computational Power.  Download the slides (PDF). Recorded at the HPC Advisory Council Switzerland Workshop on March 14, 2012.

Nvidia Shows Off First ‘Kepler’ GPUs – PCs first, Server GPU coprocessors in Q3

By Timothy Prickett Morgan • Get more from this author Graphics chip and PC and server processor wannabe Nvidia is lifting the skirt a bit on its next-generation “Kepler” graphics processing units today as it starts talking about the feeds and speeds of its new GeForce graphics cards for desktop and notebook PCs. As Nvidia co-founder and […]

Video: Cray and the Changing Landscape of High Performance Computing

httpv://www.youtube.com/watch?v=6-IpulBBGPM In this video, Peter Ungaro, president and CEO of Cray Inc. presents: The Changing Landscape of High Performance Computing at NCSA on March 14. This talk was co-sponsored by NCSA and the Illinois Parallel Computing Institute. A Tip of the Hat goes to HPC Guru for pointing us to this story.

Dell Taps Mellanox to Accelerate Blade Performance

This week Mellanox announced that Dell is using the company’s 10/40 Gigabit Ethernet and FDR InfiniBand solutions to up performance and flexibility of M620 blade servers. Mellanox solutions are the core interconnect across data centers integrating servers, storage systems and users with industry leading performance. With today’s announcement, customers can order new Dell M620 blade […]

Video: Fat-trees and Dragonflies – A Perspective on Topologies

httpv://www.youtube.com/watch?v=RnhUcW8sKXE In this video, Sven-Arne Reinemo from Simula Laboratory in Norway presents: Fat-trees and Dragonflies – A Perspective on Topologies. Download the slides (PDF). Recorded at the HPC Advisory Council Switzerland Workshop on March 15, 2012.

Xyratex Launches Most-Dense Storage Platform on the Planet

Today Xyratex announced  a new ultra-dense storage platform for OEM providers focused on scale-out storage markets for the cloud and datacenters. With up to 2 PB in a single rack, the OneStor 2584 is designed to meet data growth and management challenges with a flexible and scalable architecture. IDC predicts data storage usage will continue to […]