December 14, 2007
SAN JOSE, Calif., Dec. 10 -- Xilinx, Inc., the world's leading provider of programmable solutions, today announced that an SGI RASC (Reconfigurable Application Specific Computing) enabled SGI Altix system from Silicon Graphics, Inc. (SGI), featuring Xilinx Virtex-4 high performance FPGAs, can accelerate the Blast-n (Basic Local Alignment Search Tool for nucleotides) bioinformatics application by more than 900 times compared to a traditional cluster.
"The performance of Xilinx accelerated bioinformatics is truly astonishing and the price-performance of our RC100 based Blast-n appliance is extending the utility of this tool to many more customers and applications," said Bill Mannel, SGI director of marketing for servers. "We are seeing tremendous interest from commercial customers anxious to exploit this technology to competitive advantage."
The benchmark test ran a standard Blast-n query to match 25 nucleotide base pairs against 600,000 queries. This application required approximately 3 weeks to complete on a 68-node AMD Opteron cluster compared to less than 33 minutes for the Virtex-4 FPGA accelerated SGI RASC platform: a total speed improvement of more than 900 times.(1)
The product used to set the Blast-n performance record was an SGI Altix 4700 system configured as a turn-key bioinformatics appliance with 64 Intel Itanium 2 processors and 35 RC100 RASC blades. The completed system fits into a single rack and runs a Mitrionics developed Blast-n engine to transparently accelerate a customer's Blast-n applications using the RC100 RASC blades. Each RC100 is tightly integrated into SGI NUMAflex architecture and features two Xilinx Virtex-4 LX200 FPGAs and 10 banks of local scratchpad memory, providing a total of 70 FPGAs and 840 GB per second of local memory bandwidth in the benchmarked configuration.(2)
Now in its 4th generation, Xilinx enabled RASC technology can scale performance across a broad range of data intensive algorithms such as those used in Blast-n, the world's most widely used bioinformatics application. Additional applications appropriate for RASC acceleration include oil and gas exploration, defense and intelligence, financial analytics, medical imaging and broadcast media encoding.
"The SGI Blast-n appliance exemplifies the value FPGAs can bring to high performance computing and validates our investments in this market," said Ivo Bolsens, CTO of Xilinx. "With our newly announced Accelerated Computing Platform (ACP) for the Intel Front Side Bus, Xilinx is extending this value into the X86 platform space too."
Last month, Xilinx began commercial licensing of the high-performance computing industry's first FPGA-based acceleration solution to interface with the Intel Front Side Bus (FSB). Enabled by the high-performance 65nm Virtex-5 platform FPGA and Intel QuickAssist Technology, the ACP M1 licensing package supports implementations capable of full 1066MHz FSB performance. The ACP M1 licensing package is available today to system integrators for developing solutions that accelerate the performance of Intel processor-based server platforms while minimizing power consumption and total cost of ownership.
Customers and partners who wish to learn more about Xilinx in accelerated computing can contact ACP@Xilinx.com.
SGI (NASDAQ: SGIC) is a leader in high-performance computing. SGI delivers a broad range of high-performance server and storage solutions along with industry-leading professional services and support that enable its customers to overcome the challenges of complex data-intensive workflows and accelerate breakthrough discoveries, innovation and information transformation. SGI solutions help customers solve their computing challenges whether it's enhancing the quality of life through drug research, designing and manufacturing safer and more efficient cars and airplanes, studying global climate change, providing technologies for homeland security and defense, or helping enterprises manage large data. With offices worldwide, the company is headquartered in Sunnyvale, Calif., and can be found on the Web at www.sgi.com.
Xilinx, Inc. (NASDAQ: XLNX) is the worldwide leader of programmable logic solutions. Additional information about Xilinx is available at www.xilinx.com.
(1) Results compared to industry-standard Opteron processor-based system measured in internal tests. Running a released version of BLAST, SGI used a test case from Affymetrix comparing approximately 600,000 queries with a query size of 25 base pairs against the Human Unigene and Human ReSeq databases, which is representative of current top-end research in the pharmaceutical industry. Total execution time on a traditional 68-node Opteron-based cluster would be approximately three weeks. On the SGI reconfigurable supercomputer, benchmark input data was split in 169 jobs, which were run in groups of 70, 70 and 29 FPGAs. Total wall clock time for the run was 32m:29.183s, representing a 916X speedup over the 68-node traditional cluster.
(2) The tested SGI configuration consisted of 35 dual-FPGA SGI RC100 RASC blades, a 64-processor SGI Altix 4700 with 256GB of globally addressable memory, and standard SUSE Linux Enterprise Server 10 SP1 (kernel version 220.127.116.11-0.12) running an unmodified release of RASCAL, SGI's RASC Abstraction Layer.
Source: Xilinx, Inc.
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?