Data-Driven Multithreading: A paradigm shift for High Performance Computing

Skevos Evripidou
University of Cyprus

2:15pm-3:15pm, 20 April 2015
EM G.44

Abstract

The exponential growth of the sequential processors has come to an end at the dawn of the 20th century. Parallel processing is the only way to achieve performance growth. We propose the development of parallel architectures based on data-driven scheduling. Data-driven execution enforces only a partial ordering as dictated by the true data-dependencies which is the minimum synchronization possible. This is very beneficial for parallel processing because it enables it to exploit the maximum possible parallelism. Our team has developed the Data-Driven Multithreading (DDM). DDM has been evaluated initially in the form of virtual machines which gave very good results compared to similar systems. Our latest work provides architectural support for DDM. The main contribution of this work is the development of a highly efficient hardware support for data-driven execution and its integration into a multi-core system with eight cores on a Virtex-6 FPGA. The performance evaluation has shown that the support for data-driven execution is very efficient with negligible overheads.

Keywords: Hardware, HPC, Data-FLow

Bio

Skevos (Paraskevas) Evripidou is a Professor of Computer Science at the University of Cyprus. He received his Ph.D. in 1990 from University of Southern California. From 1990 to 1994 he was on the Faculty of the department of Computer Science and Engineering of the Southern Methodist University as an Assistant Professor (tenure track). He joined the University of Cyprus is 1994 as an Associate professor. His current research interests are in Parallel Processing, Computer Architecture with emphasis on Data-Flow systems. His team has developed the Data-Driven Mutithreading (DDM).He is the co-founder and the Chair of the Cyprus Entrepreneurship Competition (CyEC) www.cye.org.cy. Dr Evripidou has participated in several projects funded by the European Union, the USA (NSF, DARPA, and DOE) and the Cyprus Research Promotion Foundation. Dr. Evripidou is a member of the IFIP Working Group 10.3, the IEEE Computer Society and ACM SIGARCH. He is also a member of the Phi Kappa Phi and Tau Beta Pi honor societies.

Host: Sven-Bodo Scholz