AWS re:Invent 2025: Your Complete Guide to High Performance Computing Sessions

AWS re:Invent 2025 returns to Las Vegas, Nevada on December 1, uniting AWS builders, customers, partners, and IT professionals from across the globe. This year’s event offers you exclusive access to compelling customer stories and insights from AWS leadership as they tackle today’s most critical challenges in high-performance computing, from accelerating scientific discovery to optimizing complex simulations at scale. Discover cutting-edge product announcements, witness live demonstrations, and gain exclusive perspectives through must-see keynotes from AWS leaders such as Matt Garman (CEO), Dr. Werner Vogels (CTO), Peter DeSantis (SVP of Utility Computing), Dave Brown (VP of Compute and Machine Learning Services), and additional industry leaders.

If you’re looking to accelerate scientific discovery, optimize complex simulations, or migrate HPC workloads to the cloud, AWS re:Invent 2025 offers 14 specialized sessions designed specifically for high-performance computing practitioners. These sessions showcase real-world implementations across industries and offer unprecedented depth in cloud-based HPC solutions for organizations looking to harness parallel computing power without traditional infrastructure constraints.

What to Expect: HPC at re:Invent 2025

This year’s HPC track covers the full spectrum of high-performance computing on AWS, including:

• AWS Parallel Computing Service (PCS) – Managed service for setting up and managing HPC clusters using Slurm scheduler

• AWS Batch – Fully managed batch computing service for running jobs at any scale
• AWS ParallelCluster – Open source cluster management tool for deploying and managing HPC clusters
• Advanced compute instances – Purpose-built compute instances optimized for tightly-coupled HPC applications
• Storage and Networking – High-performance solutions including FSx for Lustre for storage and Elastic Fabric Adapter for low-latency networking
• Industry-specific solutions – HPC applications in aerospace, automotive, life sciences, and financial services
• Cost optimization – Strategies for running HPC workloads efficiently

Complete HPC Session Catalog

Breakout Session – CMP302 | Accelerating Engineering: Cross-Industry HPC Cloud Transformations Discover how AWS HPC is transforming engineering and product development across industries. Learn how global R&D teams are using AWS HPC instances, Elastic Fabric Adapter, FSx for Lustre, and ParallelCluster to accelerate prototyping cycles, scale semiconductor design, and overcome traditional on-premises limitations. This session covers applications in automotive, semiconductor, and consumer product domains, showcasing faster design iterations, reduced time-to-market, and more efficient innovation at scale.

Breakout Session – CMP340 | HPC at Scale with AWS Parallel Computing Service (PCS) High performance computing workloads are becoming more complex and important than ever. AWS Parallel Computing Service (PCS) simplifies this by automating cluster creation, job scheduling, and resource management. Learn how PCS helps organizations improve speed, scale securely, and achieve predictable performance for demanding workloads in energy, manufacturing, and research. Hear from customers sharing experiences of building and running real-world HPC environments with PCS.

Chalk Talk – CMP201 | Architecting Solution Patterns for GPU-Accelerated HPC and AI/ML Explore architectural patterns for GPU-accelerated High-Performance Computing (HPC) and AI/ML workloads on AWS. This interactive chalk talk covers real-world solution architectures and technical considerations for deploying workloads. Learn about EC2 GPU instance families and their optimal use cases, discuss architectural trade-offs, and explore cost optimization strategies using different provisioning models to design resilient, scalable, and cost-effective GPU-accelerated solutions.

Chalk Talk – CMP308 | AWS Tools: Automate Simulation Jobs Recovery & Maximize Cost Savings Discover how to harness the power of AWS Spot instances for HPC workloads while achieving cost savings compared to On-Demand pricing. This session demonstrates innovative automation techniques for managing long-running simulations through checkpoint-restore capabilities. Learn practical implementations using smart job navigator solutions that allow stateful jobs to survive Spot interruptions, helping accelerate research outcomes while optimizing budget efficiency.

Chalk Talk – CMP309-R | Beyond the Console: Building Powerful AWS Operational Dashboards While each AWS service provides a management console, you sometimes need more operational information to determine whether the service is acting as you expect and your services or workloads are running efficiently. This session details how you can take advantage of service events or custom metrics to build an operational dashboard using AWS Batch as an example.

Chalk Talk – CMP211 | Hybrid Science: AI + Physics Simulations for Climate & Automotive Engineering This chalk talk explores architectures that couple AI with physics-based simulations across climate science, weather forecasting, and automotive engineering. Learn how numerical simulation outputs train AI models that feed into downstream simulation tools. Discover real-world patterns for computational efficiency and workflow orchestration using AWS technologies such as AWS Batch, including practical approaches for implementing hybrid pipelines.

Chalk Talk – CMP345 | Accelerate Drug Discovery R&D at Scale with AWS Learn how 19 of the top 20 pharma companies run their compute on AWS for drug discovery research and development. This session covers techniques including microscopy imaging, molecular dynamics simulation, and virtual screening, along with management and processing of petabytes of data with heterogeneous compute requirements. Discover how generative AI is being used to predict protein structures and incorporate scientific literature findings to drive scientific discovery.

Chalk Talk – CMP350 | Accelerating Vehicle Innovation: ML & HPC Best Practices Discover how Toyota is transforming product development by migrating HPC and ML workloads to AWS, in partnership with Deloitte. Learn how EC2 GPU-powered and HPC instances on AWS reduce development cycles from months to weeks while improving performance. This session demonstrates architectures that enable robotics research, complex simulations, and predictive analytics to create safer and more efficient engineering processes.

Workshop – CMP206 | HPC Made Easy: AWS PCS for AI & Scientific Computing Learn to create, configure, and use fully-managed compute environments with minimal setup time using AWS Parallel Computing Service. This hands-on workshop covers practical job submission techniques, resource optimization strategies, and experience with PCS’s intuitive interfaces. Discover new capabilities that streamline workflows for researchers and engineers looking to accelerate computation without infrastructure complexity.

Workshop – CMP323 | HPC Sustainability by Design: Optimizing for Planet and Performance Master the art of building sustainable HPC clusters on AWS that optimize both performance and environmental impact. Learn to architect energy-efficient HPC workloads using Graviton-based instances and AWS Well-Architected Framework’s Sustainability Pillar principles. This workshop includes practical examples from science and engineering domains, providing strategies to reduce carbon footprint while maintaining computational power and implementing green computing practices.

Builders’ Session – CMP406 | HPC Infrastructure for Financial Services using AWS Batch and AWS CDK This builders’ session guides experienced HPC practitioners starting their journey with AWS infrastructure as code, as well as AWS builders new to HPC and AWS Batch. Learn to implement HPC architecture and run a sample Financial Services “Risk Analysis Grid” workload using AWS Batch and the AWS Cloud Development Kit (CDK).

Lightning Talk – CMP353 | Illumina DRAGEN Pipelines on F2 Instances with Nextflow & AWS Batch With F1’s end-of-life approaching in December 2025, this lightning talk demonstrates optimal configuration of DRAGEN pipelines using Nextflow and AWS Batch on F2 instances. Learn performance optimization and monitoring strategies, practical solutions for common debugging challenges, and insights from a real-world biopharma migration case study for transitioning Illumina DRAGEN genomics pipelines.

Lightning Talk – CMP354 | End-to-End Automotive Simulation: From Design to Analysis on AWS Traditional automotive CFD workflows face challenges with complex multi-platform environments and data management. This lightning talk provides a live demonstration of an end-to-end aerodynamic simulation workflow using AWS Research and Engineering Studio, connecting Windows-based design tools with Linux HPC resources through AWS Parallel Computing Service, with data sharing via FSx for NetApp ONTAP.

Making the Most of Your re:Invent Experience

Before You Arrive

• Reserve Your Sessions Early: Many HPC sessions are already full or nearly full. Visit the Event Catalog and secure your spots now
• Download the AWS Events Mobile App: Available for iOS and Android for real-time updates, shuttle tracking, and session management
• Plan Your Route: Sessions span multiple venues (Caesars Forum, MGM, Mandalay Bay, Wynn, Venetian) – use the app’s venue maps and shuttle schedules
• Prepare for Hands-On Learning: Bring your laptop for workshops and builders’ sessions, and ensure you have AWS credentials ready

During the Conference

• Arrive Early for Popular Sessions: Chalk talks and breakout sessions fill up quickly, especially those featuring customer case studies
• Network Strategically: Chalk talks offer direct access to AWS HPC specialists and industry practitioners – come prepared with specific questions
• Attend the Keynotes: Don’t miss insights from AWS leadership including CEO Matt Garman, CTO Dr. Werner Vogels, and other executives on the future of cloud computing and HPC
• Take Advantage of Certification Discounts: All in-person attendees receive a 50% discount voucher for AWS certification exams

Maximize Your Learning

• Mix Session Types: Combine technical deep-dives (breakout sessions) with interactive discussions (chalk talks) and hands-on practice (workshops)
• Focus on Real-World Applications: Prioritize sessions featuring customer stories from Toyota, Shell, and pharmaceutical companies for practical insights
• Document Key Takeaways: Use the AWS Events app to bookmark sessions and take notes for follow-up after the conference

Whether you’re modernizing legacy HPC environments or building new cloud-native solutions, AWS re:Invent 2025’s HPC track provides the knowledge, tools, and connections to accelerate your computational journey.