Cache-Friendly implementations of transitive closure

In this paper we show cache-friendly implementations of the Floyd-Warshall algorithm for the All-Pairs ShortestPath problem. We first compare the best commercial compiler optimizations available with standard cachefriendly optimizations and a simple improvement involving a block layout, which reduces TLB misses. We show approximately 15% improvements using these optimizations. We also develop a general representation, the Unidirectional Space Time Representation, which can be used to generate cache-friendly implementations for a large class of algorithms. We show analytically and experimentally that this representation can be used to minimize level-1 and level-2 cache misses and TLB misses and therefore exhibits the best overall performance. Using this representation we show a 2x improvement in performance with respect to the compiler optimized implementation. Experiments were conducted on Pentium III, Alpha, and MIPS R12000 machines using problem sizes between 1024 and 2048 vertices....

• Type: Journal
• Year: 2006
• Where: JEA
• Authors: Michael Penner, Viktor K. Prasanna
• Keywords: Cache-friendly Implementations, Commercial Compiler Optimizations, JEA 2006, Standard Cachefriendly Optimizations