Selective victim caching: a method to improve the performance of direct-mapped caches

Victim caching was proposed by Jouppi (1990) as an approach to improve the miss rate of direct-mapped caches. This approach augments the direct-mapped main cache with a small fully-associate cache, called victim cache, that stores cache blocks evicted from the main cache as a result of replacements. The authors propose and evaluate an improvement of this scheme, called selective victim caching. In this scheme, incoming blocks into the first-level cache are placed selectively in the main cache or the victim cache by the use of a prediction scheme based on their past history of use. In addition, interchanges of blocks between the main cache and the victim cache are also performed selectively. The authors show that the scheme results in significant improvements in miss rate as well as the number of interchanges between the two caches, for both small and large caches (4 Kbytes-128 Kbytes). For example, simulations with four instruction traces from the SPEC Release 1 programs showed an average improvement of approximately 20 percent in miss rate over simple victim caching for a 16 K cache with a block size of 32 bytes; the number of blocks interchanged between the main and victim caches reduced by approximately 74 percent. Implementation of the scheme in an on-chip processor cache is described.<>