The San Diego Supercomputer Center (SDSC) at the University of California San Diego has been granted a supplemental award from the National Science Foundation (NSF) to double the number of graphic processing units, or GPUs, on its petascale-level Comet supercomputer in direct response to growing demand for GPU computing across a wide range of research domains. Under the supplemental NSF...
Read More About SDSC to Double ‘Comet’ Supercomputer’s Graphic Processor Count »

Power and energy are now recognized as first-order constraints in high-performance computing.  Optimizing performance under power and energy bounds requires coordination across not only the software stack (compilers, operating and runtime systems, job schedulers) but also coordination with cooling systems and outwards to electrical suppliers.  As we continue to move towards...
Read More About Call for Papers: The 13th Workshop on High-Performance, Power-Aware Computing in conjunction with IPDPS 2017 - Deadline: January 29, 2017 »

The Kyoto University Graduate School of Medicine determined that a dual-socket Intel Xeon E5-2699v3 (Haswell architecture) chipset delivers better performance than an NVIDIA K40 GPU using 16-bit arithmetic (which doubles GPU performance) when training deep learning neural networks for computational drug discovery using the Theano framework. Theano is a Python library that lets researchers...
Read More About Superior Performance Commits Kyoto University to CPUs Over GPUs »