Brand: NVIDIA PN: 699-2G500-0200-300 ENGINE ARCHITECTURE: VOLTA TENSOR CORES: 640 CUDA CORES: 5120 GPU CLOCK: 1246 MHZ BOOST CLOCK: 1380 MHZ DOUBLE PRECISION PERFORMANCE: 7 TFLOPS SINGLE PRECISION PERFORMANCE: 14 TFLOPS DEEP LEARNING: 112 TFLOPS MEMORY CONFIGURATION: 16GB MEMORY INTERFACE: HBM2 MEMORY INTERFACE WIDTH: 4096 BIT MEMORY BANDWIDTH: 900 GB/S ERROR CORRECTION CODE ( ECC ): YES ( ENABLED BY DEFAULT ) MAX MEMORY CLOCK: 876 MHZ INTERCONNECT BANDWIDTH: 32GB/S BUS SUPPORT: PCI-E 3.0 PHYSICAL BUS INTERFACE: PCI-E 3.0 X16 SUPPORTED TECHNOLOGIES: CUDA TECHNOLOGY, NVIDIA GPU BOOST, DIRECTCOMPUTE, OPENCL, OPENACC, VOLTA ARCHITECTURE, TENSOR CORES OPERATING SYSTEMS: MICROSOFT WINDOWS 7, WINDOWS 8, WINDOWS 8.1, WINDOWS 10, WINDOWS SERVER 2008 R2, WINDOWS SERVER 2012 R2, WINDOWS SERVER 2016, LINUX ( ENGLISH ) MAX POWER CONSUMPTION ( W ): 250W POWER CONNECTORS: ( 1 ) ONE 8 PIN POWER CONNECTOR COOLING SYSTEM: PASSIVE ( BIDIRECTIONAL AIRFLOW ) DUAL SLOT, FULL HEIGHT NVIDIA TESLA V100 VOLTA GPU ACCELERATOR 16GB HBM2 640 TENSOR CORES 5120 CUDA CORES MEMORY BANDWIDTH 900GB/S PCI-E 3.0 X16 GENERAL PURPOSE GRAPHICS PROCESSING UNIT GPGPU In addition to Nvidia 699-2G500-0200-300 , Liyuan Tech supply a wide range of other Server motherboard chip series transceivers. If you have any need or interest, please feel free to send inquiry to global@chn-liyuan.com