Penguin Computing Announces Availability of NVidia Tesla K20 on Altus and Relion GPU Servers

Penguin Computing, experts in high performance computing (HPC) solutions and an NVidia OEM partner, today announced the availability of NVidia’s Tesla K20 accelerator on Penguin’s line of GPU servers. The next-generation NVIDIA Tesla K20 GPU Accelerator has been designed for throughput and performance per watt. It features two ultra-efficient GK110 GPUs that provide up to 3X performance per watt for single precision applications than its predecessor, the M2090. The NVIDIA Tesla K20 features two GK110 GPUs with 2,880 CUDA cores each and 5GB of memory operating at 208 GB/second per GPU. These GPUs are well suited for single as well as double precision workloads, including seismic processing, signal, image and video analysis, molecular dynamics, engineering and mathematical applications. “Penguin offers a large portfolio of GPU platforms reaching from GPU enabled workstations and double GPU hybrid servers to high density solutions that accommodate up to 16 GPUs in a 4U form factor,” said Jussi Kukonen, Director of Product Management at Penguin Computing. “Due to this broad product portfolio, a growing number of applications that feature GPU support and the compelling performance benefits we have been seeing a tremendous growth in demand for hybrid solutions. With the record breaking performance the Tesla K20 has to offer we foresee that this trend will gain even more momentum.” Delivering a peak performance of over 3.5TFLOPs for single precision and over 1TFLOPs for double precision numbers the NVidia K20 is the world’s fastest, most energy efficient accelerator,” said Sumit Gupta, senior director of the Tesla business at NVIDIA. “As an expert in all things HPC and as an NVidia OEM partner Penguin Computing offers a large selection of platforms and understands how to make our leading edge GPU technology work for our joint customers.” The NVidia Tesla K20 is available immediately through Penguin Computing and will be on display at SC’12 in Penguin Computing’s booth 1217.