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.
Today AMD unveiled innovation in heterogeneous HPC at the Centre of New Technologies at the University of Warsaw. In a new cluster deployment called Orion, the Next Generation Sequencing Centre in Warsaw is powering bioinformatics research with 1.5 petaFLOPS of AMD FirePro S9150 server GPU performance.
Today Mellanox announced its ConnectX-4 Lx 10/25/40/50 Gigabit Ethernet adapter, delivering optimal cost-performance and scalable connectivity for Cloud, Web 2.0 and storage platforms. As the first adapter designed to serve as a direct replacement for commonly deployed 10 Gigabit Ethernet adapters, the ConnectX-4 Lx allows businesses to migrate to higher-performance technology as their bandwidth requirements increase without demanding an infrastructure overhaul or added operating expense.
“The ability to accurately and efficiently study the absorption spectra of large chemical systems necessitates the development of new algorithms and the use of different architectures. We have developed a highly parallelizable algorithm in order to study excited state properties with ab initio electronic structure theory. This approach has recently been implemented to take advantage of graphical processing units to further improve efficiency.”