For further information, please contact the General chair: Pierre-Emmanuel Gaillardon
Ritsumeikan University, Japan
Malleable Task Scheduling for Multi-/Many-core Platforms
Task scheduling has been one of the most important problems in parallel computing fields. Given a set of dependent tasks in the form of a directed acyclic graph, task scheduling decides the execution order of the tasks on multiple cores. Classic task scheduling assumes that each task is executed on one of the multiple cores, but many tasks in real-world applications can be executed on multiple cores in a data-parallel fashion. This talk describes task scheduling which decides not only the inter-task execution order but also the degree of intra-task parallelism simultaneously.
Hiroyuki Tomiyama received his B.E., M.E. and D.E. degrees in computer science from Kyushu University in 1994, 1996 and 1999, respectively. He was a visiting postdoctoral researcher at UC Irvine from 1999 to 2001, a researcher at ISIT from 2001 to 2003, an assistant professor at Nagoya University from 2003 to 2004, and an associate professor there from 2004 to 2010. In 2010, he joined Ritsumeikan University as a full professor. His current research focuses are on design methodologies for embedded and cyber-physical systems. He was General Co-Chair of MPSoC 2010 and General Chair of MPSoC 2013.