Collaborative Research: PPoSS: Planning: A Disciplined Approach to Scaling in the Post-Moore's Law Era

With this planning grant: (1) Leveraging their experience with the first study, the PIs will perform a more ambitious one with a greater diversity of subjects. By re-engaging as many 2010 subjects as possible, this becomes a longitudinal study capable of revealing trends not obvious in any single point-in-time study. (2) The PIs will use the aforementioned interactions to explore transitioning their foundational work to practice. The 2010 study results inspired the PIs' to produce breakthroughs in speculation, dependence handling, latency tolerance, and automatic parallelization. The 2020 study will serve as a vehicle to explore ways to transition these results to practice. (3) The PIs believe that hardware can be more domain-adept without being domain-specific. Using prior insights, they will explore hardware-software concepts that deliver top levels of performance without undue programmer burden. By testing these ideas in the context of the study, the PIs will be able to best frame the problem, to refine approaches, and to test hypotheses in the context of real needs, opportunities, and constraints. All of these activities will help ensure that future work in scalable systems will have the greatest impact.