A QUANTITATIVE-EVALUATION OF CACHE TYPES FOR HIGH-PERFORMANCE COMPUTER-SYSTEMS

Parallel accesses to the table lookaside buffer (TLB) and cache array are crucial for high-performance computer systems, and the choice of c ache types is one of the most important factors affecting cache perfor mance. In this paper we classify caches according to both index and ta g. Since both index and tag could be either virtual (V) or real (R), o ur classification results in four combinations or cache types. The rea l address caches with virtual tags for high-performance computer syste ms in our study are prediction-based, since index bins are generated f rom a small array and predictions could be false. As a result, we also discuss and evaluate real address MRU caches with real tags, and prop ose virtually indexed MRU caches with real tags. Each of the four cach e types and MRU caches are discussed and evaluated using trace-driven simulation. Our results show that a virtually indexed MRU cache with r eal tags is a good choice for high-performance computer systems.