Al Geist from ORNL looks ahead at how we can reach Exascale in the next decade. As described so well by Geist, this road will be lined with a series of significant challenges including the programming model, power efficiency, scalability, resilience, and data movement.
National laboratory–industry–university partnerships will be extremely important to meeting the aggressive goals of the Exascale Roadmap. Fundamentally, this approach represents a shift from simply procuring and operating large scale systems. The organizations involved in the Exascale Roadmap, vendors and researchers alike, will actively engage in the development and co-design of advanced architectures and algorithms that address the nation’s critical problems in energy, climate, security, and health.