Neuromorphic Computing for Accelerating Scientific Discovery
DE-FOA-0002352
Grants.gov  https://www.grants.gov/web/grants/view-opportunity.html?oppId=326661

Required Preapplication due 6/3/2020

This program call is focused on high-risk, high-reward basic research to explore how neuromorphic computing could address emerging scientific computational challenges in energy efficient computing primarily. The DOE SC program in Advanced Scientific Computing Research (ASCR) hereby announces its interest in proposals for basic research that significantly advances Neuromorphic Computing as a brain-inspired, energy-efficient tool for scientific discovery. DOE has the responsibility to address the energy, environmental and nuclear security challenges that face our nation. SC’s mission is the delivery of scientific discoveries and major scientific tools to transform our understanding of nature and to advance the energy, economic, and national security of the United States.

In the post-exascale computing timeframe, scientific progress will be predicated on our ability to create, transfer, and process large complex datasets from extreme scale simulations, experiments, and/or observational facilities. Scientific computing is facing multiple challenges i.e. high energy usage, memory, concurrency, parallelism, heterogeneity, input/output, storage, retrieval, fault tolerance, etc. Currently, high performance computing systems consume in the order of megawatts of power, and energy consumptions levels have been kept manageable through the usage of heterogeneous architectures. On the other hand, the energy consumption of the human brain is in the order of watts, or about six orders of magnitude more energy efficient than current state-of-the-art HPC systems.