Cray CS-Storm GPU-Optimized Cluster Supercomputer
As part of the Cray® CS™ cluster supercomputer series, Cray offers the CS-Storm™ accelerated compute cluster, an accelerator-optimized system that consists of multiple high-density multi-GPU server nodes, designed for massively parallel computing workloads. The CS-Storm air-cooled platform delivers the promise of dense accelerated computing of a GPU supercomputer without compromising system performance.
The CS-Storm system is well suited for HPC workloads in the defense, oil and gas, media and entertainment, life sciences and business intelligence sectors. Typical applications include cybersecurity, geospatial intelligence, pattern recognition, seismic processing, rendering and machine learning.
Machine Learning and Deep Learning
The CS-Storm cluster also excels at emerging machine learning use cases in which the computational power of a GPU is rapidly increasing the use of deep learning and accelerated analytics.
Each Cray CS-Storm cluster rack can hold up to 22 2U rackmount CS-Storm server nodes. Each of these server nodes integrates eight NVIDIA® Tesla® GPU accelerators, delivering up to 827 peak GPU teraflops of compute performance in one 48U rack. The system is available with a choice of NVIDIA Tesla M40, K40, K80 or PCIe P100 accelerators covering a range of single- and double-precision scientific, research and industrial use cases. Additionally, Cray’s HPC software stack includes tools that are customizable to work with most open-source and commercial compilers, schedulers and libraries. .