Posted on Monday, June 01 2009 @ 3:52 CEST by Thomas De Maesschalck
NVIDIA announced a GPGPU 1U rack-mount server from Supermicro with NVIDIA Tesla GPUs delivers 12 times the performance of a traditional quad-core CPU-based 1U server:
NVIDIA Corporation and Supermicro today announced the immediate availability of a new class of server that combines massively parallel NVIDIA(R) Tesla(TM) GPUs with multi-core CPUs in a single 1U rack-mount server. This unique configuration delivers 12 times the performance of a traditional quad-core CPU-based 1U server.
Supermicro is first to market and is demonstrating the NVIDIA Tesla-based SuperServer 6016T-GF-TM2 at Computex 2009 in Taiwan this week (booth # N806, Nangang Upper Exhibition Hall).
"Our new Tesla GPU-based SuperServer 6016T-GF Series delivers a much higher performance-per-watt and per-rack than any other 1U solution in the market today," stated Don Clegg, Vice President of Marketing, Supermicro. "This 2-Teraflop SuperServer meets the most demanding enterprise data center requirements for reliability and manageability."
Based on the same award-winning NVIDIA CUDA(TM) architecture as other Tesla GPU-based solutions, these new servers offer unprecedented levels of integration and reliability for enterprise-class users. In addition, the GPU can be tied directly into the remote monitoring capabilities of the server for significantly improved system level management.
Petrobras, the leading Brazilian International Energy company, recently spoke about its reliance on Tesla GPUs to increase the performance of their seismic codes. Petrobras has recently invested in a GPU-based cluster consisting of 190 NVIDIA Tesla GPU Computing processors for scientific parallel processing, with transformative results:
"With our GPU cluster we are getting performance improvements of 5x to 20x over our traditional multi-core CPU-based cluster," said Neiva Zago, Geophysical Technology Manager, Petrobras. "We expect that the continued use of GPUs in our business will result in significant reduction in processing time as well as savings in power consumption and datacenter floor space."
Petrobras expects scalable increases in GPU performance will continue as they build out their datacenter to deliver more than 400 TFLOPS.
