14 DAY TRIAL // Santa Clara, California-based NVIDIA has just announced the introduction of a new Tesla GPU Preconfigured Cluster, specifically designed to boost the performance and minimize the power draw inside today's data center systems. According to the chip maker, its new Tesla-based cluster can enable researchers and IT managers to lower the total cost of ownership for their servers through an energy-efficient solution capable of providing up to 30 times the performance of CPU-only based solutions, according to NVIDIA.
“There are 15 to 20 million engineers, scientists and researchers around the world struggling for time on supercomputers, which has led to a huge pent-up demand for computation,” said Andy Keane, general manager of the Tesla business at NVIDIA. “With the launch of the Tesla Preconfigured Cluster, every one of them can easily deploy a GPU-powered supercomputing cluster that dramatically reduces their power consumption while still advancing the pace of their work.”
According to the specification sheet of NVIDIA's Tesla S1070, customers will be able to take advantage of the performance delivered by 4 Tesla GPUs, which account for 960 stream processors with a core speed ranging from 1.296GHz to 1.44GHz. This solution is capable of delivering a peak single Precision floating point of between 3.73 and 4.14 TFlops, while the double Precision floating point comes at between 311 and 345 GFlops. Additional features of NVIDIA's Tesla S1070 include a total of 16GB of 512-bit memory and a maximum power consumption of 800W.
When combined with NVIDIA's CUDA technology, the many-core architecture inside the Tesla S1070 can deliver the necessary performance required by applications as computational fluid dynamics, molecular dynamics, seismic processing and financial computing. Despite no word on pricing, the chip maker says that a number of its partners, including Microway, HPC Technologies, Silicon Mechanics and more, are currently offering Tesla Preconfigured Clusters.