Amazon Web Services announced P3 instances, the next generation of Amazon Elastic Compute Cloud (Amazon EC2) GPU instances designed for compute-intensive applications that require massive parallel floating point performance, including machine learning, computational fluid dynamics, computational finance, seismic analysis, molecular modeling, genomics, and autonomous vehicle systems.
P3 instances allow customers to build and deploy advanced applications with up to 14 times better performance than previous-generation Amazon EC2 GPU compute instances, and reduce training of machine learning applications from days to hours. With up to eight NVIDIA Tesla V100 GPUs, P3 instances provide up to one petaflop of mixed-precision, 125 teraflops of single-precision, and 62 teraflops of double-precision floating point performance, as well as a 300 GB/s second-generation NVIDIA NVLink interconnect that enables high-speed, low-latency GPU-to-GPU communication. P3 instances also feature up to 64 vCPUs based on custom Intel Xeon E5 (Broadwell) processors, 488 GB of DRAM, and 25 Gbps of dedicated aggregate network bandwidth using the Elastic Network Adapter (ENA).
“When we launched our P2 instances last year, we couldn’t believe how quickly people adopted them,” said Matt Garman, Vice President of Amazon EC2. “Most of the machine learning in the cloud today is done on P2 instances, yet customers continue to be hungry for more powerful instances. By offering up to 14 times better performance than P2 instances, P3 instances will significantly reduce the time involved in training machine learning models, providing agility for developers to experiment, and optimizing machine learning without requiring large investments in on-premises GPU clusters. In addition, high performance computing applications will benefit from up to 2.7 times improvement in double-precision floating point performance.”
AWS Deep Learning Machine Images (AMIs) are available in AWS Marketplace to help customers get started within minutes. The Deep Learning AMI comes preinstalled with the latest releases of Apache MXNet, Caffe2 and TensorFlow with support for Tesla V100 GPUs, and will be updated to support P3 instances with other machine learning frameworks such as Microsoft Cognitive Toolkit and PyTorch as soon as these frameworks release support for Tesla V100 GPUs. Customers can also use the NVIDIA Volta Deep Learning AMI that integrates deep learning framework containers from NVIDIA GPU Cloud, or start with AMIs for Amazon Linux, Ubuntu 16.04, Windows Server 2012 R2, or Windows Server 2016.