November 03, 2010
NASA's Goddard Space Flight Center and Lawrence Livermore National Laboratory to join demonstration at SC10
SPRINGFIELD, Ohio, Nov. 3 -- At the Supercomputing 2010 (SC10) conference, Avetec's DICE program will demonstrate through collaboration with government, academic and HPC vendors the performance and effectiveness of applications and file systems utilizing InfiniBand over the, real-world wide-area networks (WAN). The Research Sandbox (System Area Network Demonstration) project -- outside the controlled lab environment -- will provide insight for the high performance computing (HPC) community into how such technologies can be applied in real-world settings. The DICE program will share results in real time through its Twitter account @diceprogram during SC10, November 16-18, 2010.
"The entire community has gotten behind this collaboration because it addresses the one of the most pressing challenges across all sectors of HPC, and the best way to accelerate discovery is to test across different entities," said Roger Panton, executive director of the DICE program.
The project will test encapsulated and encrypted InfiniBand data movement between high performance computing clusters. The demonstration will put to the test 16 leading HPC solutions, which have been embedded in the DICE test bed -- a geographically distributed test environment -- to evaluate and verify the performance and effectiveness of applications and file systems using InfiniBand over the WAN.
The demonstration will connect from SCinet at SC10 in New Orleans to Avetec's Springfield, Ohio facility; NASA Goddard Space Flight Center in Greenbelt, Maryland; and Lawrence Livermore National Laboratory in Livermore, California.
The vendors involved in the demonstration include the following:
Data Direct Networks
Hitachi Data Systems
In addition, the following research teams have contributed expertise and resources to the Research Sandbox project:
Lawrence Livermore National Laboratory
NASA Goddard Space Flight Center
Sandia National Lab
The Ohio State University (D.K. Panda)
The community can expect to see preliminary reports through the DICE program's Twitter log about the following aspects of the Research Sandbox experiment:
"Many organizations' data repositories are growing exponentially, and most of us need to move data efficiently and reliably to geographically disperse locations," said Dan Duffy, lead HPC systems architect at Goddard's NASA Center for Climate Simulation. "This project will leverage the interoperability of a number of products to demonstrate the movement of data, both encrypted and unencrypted, at extremely high throughputs. The goal is to maximize performance and maintain reliability."
To view the complete configuration of the Research Sandbox, go to www.diceprogram.org/sandbox.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?