November 15, 2004
At SC04, staff of the Pittsburgh Supercomputing Center (PSC) have demonstrated a 40-Gigabit per second (Gbps) link.
With two next-generation Cisco CRS-1 routing systems, fitted with OC-768 interfaces, PSC's link shows that real-world data transmission at 40 Gbps is now attainable over a single light wave, known as a lambda. OC-192 is the current standard in high-performance bandwidth, which allows 10 Gbps on each lambda. Existing 40 Gbps networks use four OC-192s. One OC-768 will support the same bandwidth as four OC-192s.
"This is the next step forward in networking interface progression," said PSC director of networking Wendy Huntoon. "We're demonstrating PSC's ability to fully utilize this bandwidth with real application data."
The demonstration shows the ability of PSC's "Ultimate Integration" technology to tie together leading-edge resources via the network. A routine supercomputing application (a Laplace solver) running on LeMieux at the PSC machine room in Monroeville, Pa uses the 40 Gbps link to write checkpoint data on a Linux server in the PSC booth.
The demonstration relies on a PSC-created transparent protocol that couples the application, running on LeMieux, to the network. These connections are operated without the necessity of high-level network knowledge by a researcher running the application and facilitate remote storage of full checkpoint data at high-performance data rates. The remote storage infrastructure appears no different to the researcher than using PSC's own local storage.
This demonstration is also the first time 40 Gbps bandwidth has been achieved over SCinet, the on-site network of the annual Supercomputing conference.
Further technical details are available at the PSC Booth at SC 2004.
The Pittsburgh Supercomputing Center is a joint effort of Carnegie Mellon University and the University of Pittsburgh together with Westinghouse Electric Company. It was established in 1986 and is supported by several federal agencies, the Commonwealth of Pennsylvania and private industry.
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?