Non-Stationary Traffic Modeling and Analysis in Network-on-Chip Based Chip Multi-Processors

Tuesday May 10, 2011
Hamerschlag Hall D-210
4:00 pm

Paul Bogdan
Electrical and Computer Engineering at Carnegie Mellon University

Abstract

Networks-on-chip (NoCs) have been proposed as a scalable solution to solving the communication problem in multicore systems. Although the queuing-based approaches have been traditionally used for performance analysis purposes, they cannot properly account for many of the traffic characteristics (e.g., non-stationary, self-similarity) that are crucial for multicore platform design when communication happens via the NoC approach. To overcome this limitation, we propose a statistical physics approach to analyze the traffic dynamics in multicore systems and show how the non-stationary effects of the NoC workload can be effectively captured; this is of fundamental significance for re-thinking the very basis of multicore systems design. Using a wide set of real application traces, we demonstrate the need for a multi-fractal approach and analyze various packet arrival properties accordingly. We have also investigated the effect of packet injection rate and the data packet sizes on the multi-fractal spectrum of NoC traffic. As a case study, we show the benefits of our multi-fractal approach in estimating the probability of missing deadlines in packet scheduling for chip multiprocessors (CMPs). These findings open new research directions into NoC optimization which require accurate time- and space-dependent traffic models.

Bio

Paul Bogdan received his BSc. in Automatic Control and Computer Science in 2004 from the “Politehnica” University of Bucharest, where he was involved in research on control theory and dynamical systems. He is currently pursuing the Ph.D. degree in electrical and computer engineering at Carnegie Mellon University, Pittsburgh, PA. His research interests include performance analysis and alternative communication paradigms for multicore systems, modeling and analysis of bio-inspired computing and applications of condensed matter physics to real world phenomena.


Back to the seminar page