NVIDIA TESLA P40

  • GPU Memory: 24 GB
  • CUDA Cores: 3840
  • Single-Precision Performance: 12 TeraFLOPS
  • System Interface: x16 PCIe Gen3

About

The new NVIDIA® Tesla® P40 accelerator is engineered to deliver the highest throughput for scale-up servers, where performance matters most. NVIDIA® Tesla® P40 has 3840 CUDA cores with a peak FP32 throughput of 12 TeraFLOP/s, and like it’s little brother P4, P40 also accelerates INT8 vector dot products (IDP2A/IDP4A instructions), with a peak throughput of 47.0 INT8 TOP/s.

Specification

Specifications
GPU Architecture
NVIDIA Pascal
CUDA Parallel Processing cores
3840
Single-Precision Performance
12 TeraFLOPS
Interger Operations [INT8]
47 TOPS* (TeraOperations per Second
Enhanced Programmability with Page Migration Engine
Yes
GPU Memory
24GB
Memory Bandwidth
346 GB/s
System Interface
PCI Express 3.0 x16
Max Power Consumption
250 W
ECC Protection
Yes
Server-Optimized for Data Center Deployment
Yes
Form Factor
4.4” H x 10.5” L, Dual Slot, Full Height
Hardware-Accelerated Video Engine
1x Decode Engine, 2x Encode Engine

Our Partners

Sign up with our newsletter to follow the latest trends in server technology