Escondido, CA – February 16, 2016 - One Stop Systems, Inc. (OSS), a leading provider of next-generation GPU-based clusters for high performance computing applications, today announced that it has achieved Preferred OEM status in the NVIDIA Partner Network (NPN).
Katie Garrison discusses The GPUltima, a petaflop in a rack of networked GPUs and how it can be used to accelerate deep learning applications.This article is published on insideHPC.
To read the article, click here.
One Stop Systems exhibited at Rice University Oil and Gas for High Performance Computing on March 2nd and 3rd, 2016. We showcased our newest HPC Oil and Gas product, The GPUltima, which is a petaflop in a rack networked GPU cluster.
About 2016 Oil and Gas HPC Workshop
• 1 PetaFlop computing in a single rack
• 10X more cores, 94% less power, 96% less space
• Upgradeable using M.2 PCIe Flash modules
• Designed to meet the Open Compute Project (OCP) Open Rack Standards
• Provides 84 Teraflops performance
• 512GB of GPU memory
• Dual Altera Arria 10 1150 GX FPGA with up to 3 TFlops
• Up to 128 GBytes DDR4 Memory & 4 GBytes of Micron HMC Memory
• Up to 290 GBytes/s Memory Bandwidth
• PCIex16 Gen 3 Host Connection