P2 GPU-backed Instances Officially Launched By AWS
Amazon Web Services, Inc. (AWS) is a giant when it comes to public cloud. The company has officially launched the P2 GPU-backed Instances that we talked about a few weeks back. As the name clearly indicated, these are the virtual machine instances that are backed by graphics processing units.
According to AWS chief evangelist Jeff Barr:
These instances were designed to chew through tough, large-scale machine learning, deep learning, computational fluid dynamics (CFD), seismic analysis, molecular modeling, genomics, and computational finance workloads,”
According to the company, P2 GPU-backed Instances will allow them to compete with the competition like Microsoft Azure and IBM SoftLayer. These use eight Nvidia Tesla K80 GPUs. Each of which consists of two Nvidia GK210 GPUs.
“Each GPU provides 12GB of memory (accessible via 240 GB/second of memory bandwidth), and 2,496 parallel processing cores.
P2 GPU-backed Instances are available for Northern Virginia, Oregon, and Ireland data center regions. AWS also announced deep learning Amazon machine image. You can check that out as well. This also comes with Caffe, MXNet, TensorFlow, Theano, and Torch open-source deep learning frameworks installed, so you need not worry about getting things up and running.
P2 GPU-backed Instances is a big step forward for Amazon and it will help them out in keeping pace with the competition.