Nvidia on Tuesday outlined its Tesla P100 GPU, designed for accelerating data centers, a supercomputer for artificial intelligence called DGX-1 and efforts in everything from autonomous vehicles to virtual reality simulations.
At its GPU Technology Conference, Nvidia said its Tesla P100 GPU will land in servers in the first quarter of 2017. The Tesla platform is being used for high performance computing, which is being adopted more for enterprise analytics workloads.
The Tesla P100 improves neural network training performance, has better interconnects to scale and has improvements for big data workloads as well as new artificial intelligence algorithms. The company also updated its software developer kit.
As for the DGX-1, Nvidia said that the system is the first deep learning supercomputer that's built for artificial intelligence. The supercomputer is meant to be turnkey and match the computing power of 250 x86 servers.
Nvidia claims that the DGX-1 will enable researchers and data scientists to better use GPU. The DGX-1 includes Nvidia's GPU training system, deep learning software as well as libraries to design neural networks. The specs include:
- Up to 170 teraflops of half-precision (FP16) peak performance
- Eight Tesla P100 GPU accelerators, 16GB memory per GPU
- NVLink Hybrid Cube Mesh
- 7TB SSD DL Cache
- Dual 10GbE, Quad InfiniBand 100Gb networking
- 3U - 3200W
Nvidia added that the DGX-1 will be available in June from Nvidia and systems integrators.
Separately, Nvidia said it would partner with Massachusetts General Hospital to advance artificial intelligence and its ability to treat diseases.
Meanwhile, Nvidia outlined its Drive PX 2 platform, which is designed to be the brain of autonomous vehicles. Drive PX 2 is designed to be paired with Tesla GPUs in the data center for better mapping and deep learning. The company also added tools for virtual reality and workplace applications.