Supermicro and NVIDIA have announced a new line of server-based machines with integrated Tesla GPUs. The Supermicro SuperServer 6016T-GF-TM2 is a single, 1U chassis with an integrated NVIDIA Tesla GPU. The new server line is marketed towards those looking to make use of NVIDIA’s CUDA programming paradigm designed for massively parallel computing on their GPUs.
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.”
This is a huge departure from the traditional footprint of GPU-integrated servers. Usually, one is required to utilize either very large nodes supporting full-height, double width cards, an NVIDIA Quadro-plex unit or an NVIDIA Tesla 1U breakout chassis. For more info, read the full article here.