Location: Merz Court M.4.13 (CPD)
Date/Time: 21st December 2010, 15:00
In this talk we present a methodology to develop efficient and deadlock free routing algorithms for Network-on-Chip (NoC) platforms which are specialized for an application or a set of concurrent applications. The proposed methodology, called Application Specific Routing Algorithm (APSRA), exploits the application specific information regarding pairs of cores which communicate and other pairs which never communicate in the NoC platform to maximize communication adaptivity and performance. The methodology also exploits the known information regarding concurrency/non-concurrency of communication transactions among cores for the same purpose. The methodology does not require virtual channels to guarantee deadlock freedom. We demonstrate, through analysis of adaptivity as well as simulation based evaluation of latency and throughput, that algorithms produced by the proposed methodology give significantly higher performance as compared to other deadlock free algorithms for both homogeneous as well as heterogeneous 2D mesh topology NoC systems. Since APSRA methodology is topology agnostic, the most appropriate general implementation of the routing function within the router is using a table. A table-based implementation of the router is costlier as compared to an algorithm-based implementation. We also propose a technique to compress routing table in a mesh topology NoC to very small sizes with very little negative effect on routing adaptivity.
published on: 21st November 2010