Prefetching, Informed Caching and Broadcast Disks.

Techniques based on stochastic modeling and compilation analysis have been previously proposed for speculating page reference sequences and driving data prefetching [42,9]. Patterson et al. apply a cost-benefit analysis based on application-disclosed access patterns for more effective buffer allocation [31]. By comparing the value of block caching to the cost of block prefetching they adapt caching policies to running applications needs. Acharya et al. propose a server architecture for repetitive transmission of data over a broadcast channel [1]. The frequency of transmitted data is determined by data popularity across the served client population.