Simulating and Estimating the Behavior of a Neuromorphic Co-Processor. In: PMES'17 Proceedings of the Second International Workshop on Post Moores Era Supercomputing

Conference · Sun Jan 01 00:00:00 EST 2017

DOI:https://doi.org/10.1145/3149526.3149529· OSTI ID:1567652

Schuman, Catherine D. ^[1]; Pooser, Raphael ^[1]; Mintz, Tiffany ^[1]; Adnan, Md Musabbir ^[2]; Rose, Garrett S. ^[2]; Ku, Bon Woong ^[3]; Lim, Sung Kyu ^[3]

---

1. Oak Ridge National Laboratory, Oak Ridge, Tennessee
2. University of Tennessee, Knoxville, Tennessee
3. Georgia Institute of Technology, Atlanta, Georgia

Neuromorphic computing represents one technology likely to be incorporated into future supercomputers. In this work, we present initial results on a potential neuromorphic co-processor, including a preliminary device design that includes memristors, estimates on energy usage for the co-processor, and performance of an on-line learning mechanism. We also present a high-level co-processor simulator used to estimate the performance of the neuromorphic co-processor on real applications. We discuss future use-cases of a potential neuromorphic co-processor in the supercomputing environment, including as an accelerator for supervised learning and for unsupervised, on-line learning tasks. Finally, we discuss plans for future work.

• Research Organization: Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)
• Sponsoring Organization: USDOE Office of Science (SC)
• DOE Contract Number: AC05-00OR22725
• OSTI ID: 1567652
• Resource Relation: Conference: Second International Workshop on Post Moores Era Supercomputing, Denver, CO, USA, November 12 - 17, 2017
• Country of Publication: United States
• Language: English