A Taxonomy of Data Prefetching Mechanisms

Abstract

Data prefetching has been considered an effective way to mask data access latency caused by cache misses and to bridge the performance gap between processor and memory. With hardware and/or software support, data prefetching brings data closer to a processor before it is actually needed. Many prefetching techniques have been proposed in the last few years to reduce data access latency by taking advantage of multi-core architectures. In this paper, we propose a taxonomy that classifies various design concerns in developing a prefetching strategy. We discuss various prefetching strategies and issues that have to be considered in designing a prefetching strategy for multi-core processors.