Evolutionary Prefetching and Caching in an Independent Storage Units Model

Modern applications demand support for a large number of clients and require large scale storage subsystems. This paper presents a theoretical model of prefetching and caching of storage objects under a parallel storage units architecture. The storage objects are defined as variable sized data blocks and a specific cache area is reserved for data prefetching and caching. An evolutionary algorithm is proposed for identifying the storage objects to be prefetched and cached. The storage object prefetching approach is experimented under certain artificial workloads of requests for a set of storage units and has shown significant performance improvement with respect to request service times, as well as cache and byte hit ratios.

• Type: Conference
• Year: 2000
• Where: ADVIS
• Authors: Athena Vakali
• Keywords: ADVIS 2000, Information Management, Scale Storage Subsystems, Storage Object, Storage Units