Moderated by Susan Wu, Outbound Product Manager, Google

The recent explosion of AI has created unprecedented demand for cloud resources to train and run these workloads, with the goal of achieving faster business value through AI but sacrificing operational and energy efficiency.

Learn about the ways to optimize your AI workloads on Kubernetes for TCO, performance and sustainability:

• Clayton Coleman, Distinguished Engineer, Google, on making Kubernetes a simpler and more efficient platform for accelerators, high-scale frameworks, and more directly supporting key ML workloads like inference and training.
• Peter Pouliout, Principal Engineer, Ampere Computing on using ARM for AI Inferencing to enhance performance while addressing their ESG goals for greater sustainability.
• Ricardo Rocha, Computing Engineer, CERN on using GPU concurrency techniques, enabling higher GPU utilization and performance.
• Lu Qiu, AI Platform Tech Lead, on using advanced data management approaches, focusing on innovative caching techniques to reduce inter-region data transfer costs while addressing data locality.