• Caractéristiques 

  MEMORY SIZE	12 or 16 GB CoWoS HBM2 with ECC
  MEMORY BUS WIDTH	4096-bit
  MEMORY BANDWIDTH	12 GB at 549 GB/s 16 GB at 732 GB/s
  CUDA CORES	3584
  PEAK DOUBLE-PRECISION FLOATING POINT PERFORMANCE	4,7 Tflops (GPU Boost Clocks)
  PEAK SINGLE PRECISION FLOATING POINT PERFORMANCE	9,3 Tflops (GPU Boost Clocks)
  PEAK HALF-PRECISION FLOATING POINT PERFORMANCE	18,7 Tflops (GPU Boost Clocks)
  MEMORY INTERFACE	PCI Express 3.0 x16
  MAX POWER CONSUMPTION	250 W
  THERMAL SOLUTION	passive Heatsink
  FORM FACTOR	111,15 mm (H) x 267,7 mm (L) Dual Slot, Full Height
  DISPLAY CONECTORS	None
  POWER CONNECTORS	8-pin CPU power connector
  WEIGHT (W/O EXTENDER)	1177g
  PACKAGE CONTENT	1x Power adapter (2 x PCIe 8-pit auf single CPU 8-pin)
  PART NUMBER AND EAN	TCSP100M-12GB-PB    EAN:   3536403351786 TCSP100M-16GB-PB    EAN:   3536403351823

WORLD'S MOST ADVANCED DATA CENTER ACCELERATOR

NVIDIA Tesla P100 GPU accelerators for PCIe based servers available with 12 GB or 16 GB HBM2 memory are the most advanced ever built.

Powered by the breakthrough NVIDIA Pascal™ architecture and designed to boost throughput and save money for HPC and hyperscale data centers. The newest addition to this family, Tesla P100 for PCIe enables a single node to replace half a rack of commodity CPU nodes by delivering lightning-fast performance in a broad range of HPC applications. Handling the same workload with far fewer nodes means customers can save up to 70% in overall data center costs.

PNY provides unsurpassed service and commitment to its professional graphics customers offering: 3 year warranty, pre- and post-sales support, dedicated Quadro Field Application engineers and direct tech support hot lines.