“At ISC 2015, Numascale announced record-breaking results from a shared memory system running the McCalpin STREAM Benchmark, a synthetic benchmark program that measures sustainable memory bandwidth and the corresponding computation rate for simple vector kernels. Numascale’s cache coherent shared memory system, which was targeted for big data analytics, reached 10.06 TBytes/second for the Scale function.”
Today SGI announced that Inova Translational Medicine Institute (ITMI) has selected the company’s supercomputing solutions to enable researchers to obtain insights from its premier genomic databases. With the ability to diagnose patients with more accuracy and speed, ITMI will enable a higher level of treatment and care for the diverse population it serves.
Dr. Eng Lim Goh from SGI discusses important trends in HPC including pending changes coming to processors/accelerators, memory hierarchy, and interconnects. “SGI, the trusted leader in high performance computing, is focused on helping customers solve their most demanding business and technology challenges by delivering technical computing, Big Data analytics, cloud computing, and petascale storage solutions that accelerate time to discovery, innovation, and profitability.”
“As data explodes in volume, velocity and variety, and the processing requirements to address business challenges become more sophisticated, the line between traditional and high performance computing is blurring,” said Bill Mannel, vice president and general manager, HPC and Big Data, HP Servers. “With this alliance, we are giving customers access to the technologies and solutions as well as the intellectual property, portfolio services and engineering support needed to evolve their compute infrastructure to capitalize on a data driven environment.”
In this podcast, the Radio Free HPC team looks at how the KatRisk startup is using GPUs on the Titan supercomputer to calculate global flood maps. “KatRisk develops event-based probabilistic models to quantify portfolio aggregate losses and exceeding probability curves. Their goal is to develop models that fully correlate all sources of flood loss including explicit consideration of tropical cyclone rainfall and storm surge.”
Today Nvidia updated its GPU-accelerated deep learning software to accelerate deep learning training performance. With new releases of DIGITS and cuDNN, the new software provides significant performance enhancements to help data scientists create more accurate neural networks through faster model training and more sophisticated model design.