The Tesla P100 Is NVIDIA's New & Most Powerful Accelerator
From NVIDIA's GPU Technology Conference, the company announced today the Tesla P100 as their most advanced accelerator based upon their Pascal "GP100" GPU.
The Tesla P100 is designed for deep learning, HPC, and similar GPU computing environments. The Tesla P100 features NVIDIA's NVLink interconnect, HBM2 memory, and is manufactured on a 16nm FinFET process. This first Pascal part should blow the earlier K40 and M40 accelerators out of the water. The P100 is rated for an incredible 5304 GFLOPs for FP64. The TDP on the Tesla P100 is 300 Watts and there are 15.3 billion transistors.
Those wishing to learn more about the Tesla P100 that aren't at NVIDIA's conference can see this blog post and the P100 product page. The Pascal details from the blog post are quite exciting. Now for some desktop Pascal GPUs to hit the market later this year...
The Tesla P100 is designed for deep learning, HPC, and similar GPU computing environments. The Tesla P100 features NVIDIA's NVLink interconnect, HBM2 memory, and is manufactured on a 16nm FinFET process. This first Pascal part should blow the earlier K40 and M40 accelerators out of the water. The P100 is rated for an incredible 5304 GFLOPs for FP64. The TDP on the Tesla P100 is 300 Watts and there are 15.3 billion transistors.
Those wishing to learn more about the Tesla P100 that aren't at NVIDIA's conference can see this blog post and the P100 product page. The Pascal details from the blog post are quite exciting. Now for some desktop Pascal GPUs to hit the market later this year...
16 Comments