December 03, 2007
NEW YORK, Nov. 29 -- ASPEED Software announces dramatic application performance and system improvements from benchmarking Intel’s new 45nm Hi-k Quad-Core Intel Xeon processor 5400 series. The benchmarks were run on a Intel Xeon processor 5400 series (“Harpertown”) running at 2.66GHz-2.83GHz with 2x 6MB shared Cache and 1333MHz Front Side Bus.
“The benchmark results answer the question of whether single-threaded applications can benefit from the multi-core technologies without having to be re-written,” said Kurt Ziegler, ASPEED Software CEO. “What makes these benchmark results even more exciting to us and our customers is that our August 2006 Intel Xeon processor 5100 series ('Woodcrest') benchmarks, our November 2006 Intel Xeon 5300 processor series ('Clovertown') benchmarks and these Xeon processor 5400 processor series benchmarks were all run on the same physical system. Our engineers simply upgraded the 5100 to the 5300 by replacing the dual core with the quad core boards. Two weeks ago we upgraded the 5300 to a 5400. The ACCELLERATED single-threaded benchmark application immediately leveraged the new architecture and delivered a 2.2-fold run time reduction while consuming only 13 percent more power, measured at the plug, when compared with the original Xeon 5100 system.”
The benchmark highlights Intel’s faster chips that consume significantly less power using the same physical space and how ASPEED enables its clients’ single-threaded ACCELLERATED applications to transparently benefit from such hardware platform advances without the need to rewrite or restructure the application.
By the numbers:
In short, the Xeon processor 5400 series upgrade from the Xeon processor 5100 series represents a 220 percent run time improvement with 13 percent more power consumption. These benchmarks used the Intel C++ compiler running a 64-bit version of AMBook on Redhat Linux.
ACCELLERANT uniquely virtualizes applications to exploit multiple cores, CPUs and distributed systems such as those running in a cluster or on a grid without any tuning to ensure predictable and scalable capacity. ASPEED, whose primary focus is in capital markets, uses an AMBook (American Options Portfolio of options pricing) application as representative of a typical Wall Street run to price a portfolio of options using Monte Carlo-based analysis. The focus is on the impact of new platforms on representative applications in addition to shear performance potential. The benchmarks include varying the number of options and altering latency in the runs to characterize computational bound to I/O bound characteristics.
Powering the Most Powerful
The Quad-Core Intel Xeon processor 5400 series is the latest in a series of ACCELLERANT deployments using multi-core, multi-processors, clusters and grids, each demonstrating the performance benefits, scalability and configuration optimization. ASPEED offers an ACCELLERANT SDK that supports C, C++, C#, Java, VB.net, EXCEL, VBA, MS Access and FORTRAN.
ACCELLERANT is regularly deployed by leading organizations within the capital markets and energy verticals to optimize application performance and speed analysis to exploit multi-core, cluster, grid and other high-performance computing environments.
About ASPEED Software
ASPEED Software Corp. is a privately held, venture-funded software company based in the heart of New York City with development centers in New York and London. ASPEED’s value proposition is to enable its clients with existing single threaded and multi-threaded applications to significantly reduce response times and improve precision by leveraging multi-core systems, clusters and grids. ASPEED’s software is used in some of the industry’s leading financial institutions improving competitiveness, productivity and accuracy of the analysis. For more information, visit www.aspeed.com.
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?