Professor of Computer Science
University of Rochester in Rochester, New York
Theory and Practice of Multicore Cache Optimization
Wednesday, February 8, 2017
to 5:00 PM
DH 1070 Duncan Hall
6100 Main St
Houston, Texas, USA
Most of today's programs are run on multicore processors, and as a result they interact and interfere, especially by sharing the memory system. Locality theory is concerned with the fundamental properties of data usage. In this talk I will introduce multiple branches of the past research that culminate in our higher-order theory of locality (HOTL) [PPOPP'11,PACT'11,ASPLOS’13,USENIX’16,PPOPP’17] and its applications in shared cache analysis and optimization for both hardware and software cache [CCGrid'12/15,ICPP'15,USENIX'15].
Biography of Chen Ding:
Chen Ding a Professor of Computer Science at University of Rochester in Rochester, New York. He received a Ph.D. from Rice University, an M.S. from Michigan Tech, and a B.S. from Beijing University. His work received young investigator awards from NSF and DOE.