The new blade will offer significant improvements in the performance of applications used in such fields as oil and gas, medical imaging and bioinformatics, SGI says.
Silicon Graphics is rolling out a blade server designed to speed up compute-intensive applications in the high-performance computing space.
The RASC RC100 blade takes advantage of new FPGA (field-programmable gate array) technology from Xilinx; SGIs own servers and technology added to that can offer significant improvements in the performance of applications used in such fields as oil and gas, medical imaging and bioinformatics, said Bill Mannel, director of product marketing for Mountain View, Calif.-based SGI.
FPGAs are integrated circuits that can be programmed in the field after they are manufactured.
In the case of SGIs new blade, these can be programmed at the end users site to speed up particular applications.
Mannel said the combination of the FPGA technology with SGIs Altix servers and its NUMAflex shared memory architecture, the high-speed NUMAlink interconnect, and RASC (or Reconfigurable Application-Specific Computing) technology translates to the improvements in application performance, lower power consumption and enhanced price and performance.
Using NUMAlink with the FPGA brings greater bandwidth over using the gate array by itself, and the shared memory means faster reprogramming of the FPGA, according to SGI.
The RC100 blade uses two Xilinx Virtex 4 LX200 FPGAs and NUMAlink 4 ports and can be installed on SGIs Altix 4000 systems.
Check out eWEEK.coms for the latest news, views and analysis on servers, switches and networking protocols for the enterprise and small businesses.