Integrated Tracy Kimbrel * Pei Cao Parallel t Prefetching W. Felten I and Caching Anna R. Karlin* Kai Li$ Edward Introduction Prefetching improving and caching the are widely-used approaches for Model and algorithms model of the parallel prefetch- We consider performance of file systems. A recent the following ing problem. There is a two-level memory hierarchy: a cache of k data blocks and d disks. A program makes sequence, of refera known sequence, called a reference encesrl, rz, . . . rn to a set of m data blocks, If a reference hits in the cache, it takes one time unit. Otherwise, the missing block must be fetched from a disk, which study shows that it is important and proposed an algorithm that to integrate the two, performs well both in theory and in practice [2, I]. That study waa restricted to the case of a single disk. Here, we study integrated prefetching and caching strategies for multiple disks. The interaction between caching and prefetching is further complicated when a system has multiple disks, not only because it is possible to do multiple prefetches in parallel, but also because appropriate cache replacement strategies can alleviate the load imbalance
/lp/association-for-computing-machinery/integrated-parallel-prefetching-and-caching-CNtf0LfP3j