November 13, 2012
SALT LAKE CITY, Nov. 13 – Today at SC12, Gnodal Limited, a systems company delivering industry leading performance, port density and low power to the high-performance data center with GS-Series switches; and Chelsio, a leader in 10 Gigabit Ethernet (10GbE) Unified Wire networking adapters, announced industry leading benchmarks of MPI performance using the OSU MPI stack in a test with the Gnodal GS7200 10/40GbE switch and Chelsio T420-LL-CR adapters.
Ultra-low MPI latency is required by many commercial scientific and engineering applications, as well as by many tools used in Research and Academia. Providing such low latency using Ethernet—so far only achieved using proprietary networks—offers the benefit of supporting high-performance computing workloads along with storage and regular traffic within a standards-based converged networking environment. Additional information about the test environment and performance can be found at T4 Benchmark Open MPI OFED iWARP.
"We are observing a trend in HPC as organizations are now able to exploit standards-based Ethernet networks to deliver the superior HPC performance they desire," said Dr. John Taylor, Gnodal Vice President of Technical Marketing. "This test shows that with the combination of the Chelsio 10GbE Unified Wire server adapter and the scalable Gnodal GS-Series switch family, we are able to provide an industry leading, end-to-end, low-latency solution capable of handling the most demanding HPC applications with the added benefit of reducing the complexity of managing multiple networks."
"HPC customers are typically focused on attaining the lowest latency possible, and the combination of Gnodal switches and our Unified Wire adapters is a powerful solution that addresses their needs," said Dr. Wael Noureddine, Vice President of Technology and Product Management, Chelsio. "The T420-LL-CR adapter is proven to provide the lowest 10Gb iWARP latency in the industry, meeting the requirements of the most delay-sensitive applications, including High Performance Computing and High Frequency Trading."
Chelsio's T420-LL-CR 10GbE Unified Wire adapters leverage Chelsio's fourth-generation, high-performance, Transport Offload Engine (TOE) processor core. Chelsio's ASIC fully implements the complete TCP/IP stack, as well as higher level protocols including iWARP RDMA and iSCSI, thereby offloading protocol-processing costs from the host to the adapter. Transport-offload technology frees up server memory, bandwidth and expensive CPU cycles to improve the performance of useful applications, while preserving the reliability and scalability benefits of the familiar TCP/IP stack.
In addition to testing with 10GbE, Chelsio's performance projections for 40GbE iWARP herald even more opportunity to transition from specialized fabrics to high-speed Ethernet. This highlights a further benefit of the built-in capabilities of the Gnodal PETA ASIC in supporting native 40GbE within the same platform. More information about Chelsio's overall performance benchmarks can be found at http://www.chelsio.com/performance-benchmarks/.
About Chelsio Communications
Chelsio is a leading technology company focused on solving high performance networking and storage challenges for virtualized enterprise data centers, cloud service installations, and cluster computing environments. Now shipping its fourth-generation protocol acceleration technology, Chelsio is delivering hardware and software solutions including Unified Wire Ethernet network adapter cards, unified storage software, high performance storage gateways, unified management software, bypass cards, and other solutions focused on specialized applications. For more information about Chelsio, visit booth # 3818 at SC12.
About Gnodal
The Gnodal ASIC Ethernet switch architecture features a congestion-aware performance and workload engine that allows for ultra-low latency transmission while utilizing a dynamic, fully adaptive load-balancing mechanism to equitably arbitrate a pre-emptive pathway for large data-sets, high-computational applications and massive storage demands prevalent in HPC and Big Data environments. The 72-port, 40GbE, "fabric-in-a-box" GS0072 solution extends Gnodal's leadership in port density ToR solutions and won the best-in-class award for networking at Interop 2012 (www.bestofinterop.com/winners/).
Gnodal's high-performance network fabric delivers industry-leading speed to help reduce latencies. Gnodal's highest port density 1U and 2U ToR switches are ideally suited for deployment within co-location environments and enterprise data centers. On ingress into GS-Series switching, the initial latency is sub-150 nanoseconds (store/forward) with each subsequent Gnodal switch added to the fabric incurring only 66 nanoseconds of additional latency (cut-through). For more information about Gnodal, visit booth #4818 at SC12.
-----
Source: Gnodal
Large-scale, worldwide scientific initiatives rely on some cloud-based system to both coordinate efforts and manage computational efforts at peak times that cannot be contained within the combined in-house HPC resources. Last week at Google I/O, Brookhaven National Lab’s Sergey Panitkin discussed the role of the Google Compute Engine in providing computational support to ATLAS, a detector of high-energy particles at the Large Hadron Collider (LHC).
Read more...
The Xeon Phi coprocessor might be the new kid on the high performance block, but out of all first-rate kickers of the Intel tires, the Texas Advanced Computing Center (TACC) got the first real jab with its new top ten Stampede system.We talk with the center's Karl Schultz about the challenges of programming for Phi--but more specifically, the optimization...
Read more...
Although Horst Simon was named Deputy Director of Lawrence Berkeley National Laboratory, he maintains his strong ties to the scientific computing community as an editor of the TOP500 list and as an invited speaker at conferences.
Read more...
May 16, 2013 |
When it comes to cloud, long distances mean unacceptably high latencies. Researchers from the University of Bonn in Germany examined those latency issues of doing CFD modeling in the cloud by utilizing a common CFD and its utilization in HPC instance types including both CPU and GPU cores of Amazon EC2.
Read more...
May 15, 2013 |
Supercomputers at the Department of Energy’s National Energy Research Scientific Computing Center (NERSC) have worked on important computational problems such as collapse of the atomic state, the optimization of chemical catalysts, and now modeling popping bubbles.
Read more...
May 10, 2013 |
Program provides cash awards up to $10,000 for the best open-source end-user applications deployed on 100G network.
Read more...
May 09, 2013 |
The Japanese government has revealed its plans to best its previous K Computer efforts with what they hope will be the first exascale system...
Read more...
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
In this demonstration of SGI DMF ZeroWatt disk solution, Dr. Eng Lim Goh, SGI CTO, discusses a function of SGI DMF software to reduce costs and power consumption in an exascale (Big Data) storage datacenter.
The Cray CS300-AC cluster supercomputer offers energy efficient, air-cooled design based on modular, industry-standard platforms featuring the latest processor and network technologies and a wide range of datacenter cooling requirements.