AWS EC2 compute is evolving to meet the growing demand for agentic AI and physical AI, with customers pushing general-purpose cloud infrastructure into new territory. Twenty years on, the platform is seeing growth throughout every layer, according to Art Baudo, principal product marketing manager and head of EC2 product marketing at Amazon Web Services Inc.
Customers are now using AMD-based instances for AI inference workloads alongside traditional high-performance computing tasks. The breadth of what is moving into the cloud has surprised even longtime cloud advocates.
EC2 Compute Expansion
Art Baudo said that as performance has gone up, so has the price performance delivered to customers. With AI taking off, people are using it in multiple different spaces across the industry and across the instance types as well.
Art Baudo spoke with theCUBE’s Dave Vellante and John Furrier at the AMD Advancing AI event. They discussed how the AWS and AMD partnership has matured since the first EPYC instances launched in 2018.
Related: Meta updates Muse Spark and Facebook
AMD Partnership Depth
AWS introduced AMD EPYC processors into EC2 in 2018 and has shipped every subsequent generation, including Turin-based instances launched in 2026. The shift from multithreaded to single-threaded instances between the sixth and seventh generation delivered substantial performance gains.
Customers in AI and EDA workloads have been able to take advantage of these gains. More recently, AWS has introduced high-frequency instances that combine 5GHz clock speeds with expanded memory configurations to serve workloads that require peak compute and rapid data access.
Art Baudo said that they have tried to continuously make sure they optimize on price performance for customers. Some of their most recent instances continue to see a reduction in price performance.
AWS Nitro System
The AWS Nitro System resides beneath all of this as the architectural foundation that makes EC2 compute’s breadth and security possible. By offloading hypervisor functions to dedicated hardware and software, Nitro allows AWS to deploy instance types much faster and at greater scale while enforcing zero operator access.
Related: Cast Insights raises 4.5 million dollars
Art Baudo said the system extends naturally into AI workloads throughout the full portfolio, from AMD and Graviton instances to Trainium and Inferentia, providing customers with a consistent security baseline, regardless of which compute tier they choose.
Art Baudo said that AWS is applying the same customer-centric optimization discipline to AI workloads that it applied to general cloud cost management in 2022. The pattern parallels what AWS has done historically, making the economics work so customers buy more compute, not less.
Art Baudo said that they have been trying to help customers get more information about Spot Instances so they can take advantage of these cost-saving vehicles within their ecosystem. They show customers where instances are available and give that information to them so they can use the spot side even more.
