AWS launches next-gen GPU instances for machine learning

P4d adds Intel Cascade Lake CPUs and Nvidia GPUs for high-performance computing

AWS has launched its latest GPU-equipped instances aimed at machine learning and high-performance computing (HPC) workloads.

Called P4d, the new instances come ten years the first set of GPU instances were launched. They feature Intel Cascade Lake processors and eight of Nvidia's A100 Tensor Core GPUs. These connect via NVLink with support for Nvidia GPUDirect and offer 2.5 PetaFLOPS of floating-point performance and 320GB of high-bandwidth GPU memory.

AWS claimed that the instances offer 2.5x the deep learning performance, and up to 60% lower cost to train when compared to P3 instances.

In addition, the P4 instances include 1.1TB of system memory and 8TB of NVME-based SSD storage with up to 16 gigabytes of read throughput per second. The instances can combine over 4,000 GPUs into an on-demand EC2 UltraCluster. 

Among the use cases touted by AWS for these instances include supercomputer-scale machine learning and HPC workloads: natural language processing, object detection & classification, scene understanding, seismic analysis, weather forecasting, financial modelling, etc. 

The P4 instances are available in one size (p4d.24xlarge) and can be launched in the US East (N.Virginia) and US West (Oregon) Regions with immediate effect.

Among the companies that have already been working with the P4 instances include Toyota Research Institute (TRI), GE Healthcare and Aon.

"At TRI, we're working to build a future where everyone has the freedom to move,” said Mike Garrison, technical lead, Infrastructure Engineering at TRI.

"The previous generation P3 instances helped us reduce our time to train machine learning models from days to hours and we are looking forward to utilizing P4d instances, as the additional GPU memory and more efficient float formats will allow our machine learning team to train with more complex models at an even faster speed."

Its on-demand price will be $32.77 per hour, going down to approximately $20 per hour for one-year reserved instances, and $11.57 for three-year reserved instances.

Featured Resources

Next-generation time series: Forecasting for the real world, not the ideal world

Solve time series problems with AI

Free download

The future of productivity

Driving your business forward with Microsoft Office 365

Free download

How to plan for endpoint security against ever-evolving cyber threats

Safeguard your devices, data, and reputation

Free download

A quantitative comparison of UPS monitoring and servicing approaches across edge environments

Effective UPS fleet management

Free download

Most Popular

Best Linux distros 2021
operating systems

Best Linux distros 2021

11 Oct 2021
Cryptocurrency: Should you invest?
cryptocurrencies

Cryptocurrency: Should you invest?

27 Oct 2021
Apple MacBook Pro 15in vs Dell XPS 15: Clash of the titans
Laptops

Apple MacBook Pro 15in vs Dell XPS 15: Clash of the titans

11 Oct 2021