NVIDIA® V100
Every industry wants intelligence. Within their ever-growing oceans of data lie insights that can revolutionize entire industries. Personalized cancer therapy, predicting the next big hurricane, virtual personal assistants conversing naturally. These opportunities can become a reality when data scientists are given the tool they need to realize their life’s work.
NVIDIA® V100 is the world’s most advanced data center GPU ever built to accelerate AI, HPC, and Graphics. Powered by the latest GPU architecture, NVIDIA VoltaTM, V100 offers the performance of 100 CPUs in a single GPU—enabling data scientists, researchers, and engineers to tackle challenges that were once impossible.
NVIDIA V100 Specifications
- 5120 CUDA cores
- 640 New Tensor Cores
- 7.5 TeraFLOPS double-precision performance with NVIDIA GPU Boost
- 15 TeraFLOPS single-precision performance with NVIDIA GPU Boost
- 120 TeraFLOPS mixed-precision deep learning performance with NVIDIA GPU Boost
- 300 GB/s bi-directional interconnect bandwidth with NVIDIA NVLink
- 900 GB/s memory bandwidth with CoWoS HBM2 Stacked Memory
- 16 GB of CoWoS HBM2 Stacked Memory
- 300 Watt
NVIDIA V100 Form Factors
- NVIDIA V100 for NVLink: Ultimate performance for deep learning
- NVIDIA V100 for PCIe: Highest versatility for all workloads
Selected NVIDIA V100 Servers
Relion® XE1114GT
The Penguin Computing® Relion® XE1114GT is a 1U, 4x GPU server powered by the Intel® Xeon® Processor Scalable Family, targeting enterprise customers and highly optimized deep learning applications. Relion XE1114GT supports up to 4x PCIe GPU in a balance PCIe Topology.