Abstract: Computing at large scales has become extremely challenging due to increasing heterogeneity in both hardware and software. A positive feedback loop of more scientific insight leading to more complex solvers which in turn need more computational resources has been a continuous driver for development of more powerful platforms. The field of computer architecture is poised for more radical changes in how future platforms are likely to be designed, especially because scientific workflows themselves are growing more complex and diverse. These constraints indicate a need to fundamentally rethink the way computational science are designed and the tools and methodologies that are needed to enable these complex workflows.

In this seminar I will present a distillation of the Flash-X experience, including its design features for performance portability, and the design of its software process for sustainability, reproducibility, and for doing quality science.