Active Flash: Towards Energy-Efficient, In-Situ Data Analytics on Extreme-Scale Machines

Modern scientific discovery is increasingly driven by large-scale supercomputing simulations, followed by data analysis tasks. These data analyses are either per-formed offline, on smaller-scale clusters, or on the su-percomputer itself. Unfortunately, these techniques suf-fer from performance and energy inefficiencies due to in-creased data movement between the compute and storage subsystems. Therefore, we propose Active Flash, an in-situ scientific data analysis approach, wherein data anal-ysis is conducted on the solid-state device (SSD), where the data already resides. Our performance and energy models show that Active Flash has the potential to ad-dress many of the aforementioned concerns without de-grading HPC simulation performance. In addition, we demonstrate an Active Flash prototype built on a com-mercial SSD controller, which further reaffirms the via-bility of our proposal.