Prefetching using Markov predictors

Coterminous locality and coterminous group data prefetching on chip-multiprocessors

“…(This is similar to the miss correlation in [6,11].) In measuring the locality, up to four successors of each reference are kept based on the LRU replacement.…”

“…Prior correlation-based predictors (e.g. Markov predictor [11,6] and Global History Buffer [20]) record and use past miss correlations to predict future cache misses. However, a huge history table or FIFO is usually needed to provide decent coverage.…”

“…Traditional data prefetching based on cache miss correlations [6,11] faces serious obstacles. First, each cache miss often has several potential successive misses and prefetching multiple successors is inaccurate and expensive.…”

Last-Touch Correlated Data Streaming

“…Another proposed class of prefetchers utilizes address correlation [3,4,10,11,15,20], which promises wider applicability across a diverse spectrum of workloads because they target generalized memory access patterns. Rather than detecting patterns in data layout, these prefetchers correlate data addresses to predict future misses.…”

Data Cache Prefetching Using a Global History Buffer

“…[3][4][5][6] Figure 1a illustrates conventional We propose an alternative structure, shown in Figure 1b, for holding prefetch history. In this structure, a fixed-length FIFO table, the global history buffer (GHB) holds cache miss addresses.…”