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.
Sign up today and you will receive a free copy of our Future Focus 2026 report - the leading resource for IT decision-maker insight on priorities and investment areas in AI, security and more.
"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.
Rene Millman is a freelance writer and broadcaster who covers cybersecurity, AI, IoT, and the cloud. He also works as a contributing analyst at GigaOm and has previously worked as an analyst for Gartner covering the infrastructure market. He has made numerous television appearances to give his views and expertise on technology trends and companies that affect and shape our lives. You can follow Rene Millman on Twitter.
-
How AMD EPYC CPUs and AMD Instinct GPUs are powering the world's HPC facilitiesSponsored AMD offers powerful chipsets for AI and traditional HPC workloads, including native FP64 support
-
AWS pledges $50 billion to expand AI and HPC infrastructure for US government clientsNews The company said an extra 1.3 gigawatts of compute capacity will help government agencies advance America’s AI leadership
-
US Department of Energy’s supercomputer shopping spree continues with Solstice and EquinoxNews The new supercomputers will use Oracle and Nvidia hardware and reside at Argonne National Laboratory
-
Destination Earth: The digital twin helping to predict – and prevent – climate changeSupported How the European Commission is hoping to fight back against climate change with a perfect digital twin of our planet, powered by Finland's monstrous LUMI supercomputer
-
Supercomputing in the real worldSupported From identifying diseases more accurately to simulating climate change and nuclear arsenals, supercomputers are pushing the boundaries of what we thought possible
-
What is exascale computing? Exploring the next step in supercomputersSupported 60 years after the birth of the first supercomputers, we are entering a new era
-
Discover the six superpowers of Dell PowerEdge serverswhitepaper Transforming your data center into a generator for hero-sized innovations and ideas.
-
Enhance end-to-end data security with Microsoft SQL Server, Dell™ PowerEdge™ Servers and Windows Server 2022whitepaper How High Performance Computing (HPC) is making great ideas greater, bringing out their boundless potential, and driving innovation forward