Pierre-Yves Aquilanti

Amazon EC2 offers a broad set of accelerators in the cloud for machine learning. Whether you are using instances powered by NVIDIA GPUs, AWS Trainium, or AWS Inferentia, you may wonder how to manage the software stack.

Generative AI and large language models (LLMs) are rapidly disrupting workflows. Inference architectures for these large-scale models need to be carefully constructed to achieve optimal cost performance.

Elastic Fabric Adapter (EFA), built on the Scalable Reliable Datagram (SRD) protocol, is the foundation for scaling message passing interface (MPI) and NVIDIA Collective Communications Library–based high performance computing (HPC) and machine learning (ML) codes at AWS. This builders’ session addresses application scaling challenges and is led by an AWS EFA team principal engineer.

Ever wondered how Stable Diffusion—the deep learning, text-to-image model—was trained? Attend this session to hear AWS and Stability AI discuss best practices, obstacles, and key considerations when training large AI models at scale.

In this workshop, walk through how you can create complete HPC systems on AWS and benefit from the elasticity, flexibility, and scalability the cloud provides. Learn how the latest updates to AWS ParallelCluster make it easier than ever to create and manage completely dynamic HPC clusters on demand through the use of a new API.

In this chalk talk, learn about different types of architectures and tools that can help you manage your batch jobs on AWS. Find out architectural best practices for Spot Instance allocation strategies, instance selection, and using events to gather information on workloads.

It’s not hard to build your first HPC cluster on AWS. In this workshop, we take you through the basics in a step-by-step approach that you can easily follow along with.

Autonomous driving development requires that developers virtually drive millions of miles per day in thousands of scenarios and their “what-if” variants before deploying autonomous software on physical vehicles. Providing quick feedback to developers on how their software performs requires running those driving simulations at massive scale.