pchardware.org
Categories

NVIDIA · 2016

NVIDIA P100 GPU accelerator

The NVIDIA P100 GPU accelerator, released in 2016, is a high-performance computing accelerator based on the Tesla architecture and GP100 chip.

The NVIDIA P100 GPU accelerator is a high-performance computing (HPC) accelerator released in 2016. It is based on the Tesla architecture and features the GP100 chip built on a 16 nm process. The accelerator includes 3584 shading units, with a base clock of 1328 MHz and a boost clock of 1480 MHz. It is equipped with 16 GB of HBM2 memory on a 4096-bit bus, providing a memory bandwidth of 732 GB/s. The P100 has a board power (TDP) of 250 W and requires a single 8-pin power connector. It uses a PCIe 3.0 x16 bus interface. The launch MSRP was 5999 USD.

Specifications

Architecture unverified Tesla
GPU chip unverified GP100
Process node unverified 16 nm
Shading units unverified 3584
Base clock unverified 1328 MHz
Boost clock unverified 1480 MHz
Memory size unverified 16 GB
Memory type unverified HBM2
Memory bus unverified 4096 bit
Memory bandwidth unverified 732 GB/s
Board power (TDP) unverified 250 W
Power connectors unverified 1x 8-pin
Bus interface unverified PCIe 3.0 x16
Release date unverified 2016-04-05
Release year unverified 2016
Launch MSRP unverified 5999 USD