Page allocation to reduce access time of physical caches

作者: William L. Lunch , Michael J. Flynn , Brian K. Bray

DOI:

关键词: Associative propertyAccess timeCache coherenceCache accessLatency (engineering)Allocation algorithmCacheParallel computingComputer scienceColored

摘要: A simple modification to an operating system''s page allocation algorithm can give physically addressed caches the speed of virtually caches. Colored reduces number bits that need be translated before cache access, allowing large low-associativity indexed address translation, which latency processor. The colored also has other benefits: miss less (in general) and more uniformly, inclusion principle holds for second level with associativity. However, requires main memory partitioning, common shared virtual addresses. Simulation results show high non-uniformity rates normal allocation. Analysis demonstrates extent second-level inclusion, reduction in effective main-memory due partitioning.

参考文章(10)
John L. Hennessy, Steven A. Przybylski, Performance directed memory hierarchy design Stanford University. ,(1988)
M.D. Hill, A case for direct-mapped caches IEEE Computer. ,vol. 21, pp. 25- 40 ,(1988) , 10.1109/2.16187
Douglas W. Clark, Joel S. Emer, Performance of the VAX-11/780 translation buffer: simulation and measurement ACM Transactions on Computer Systems. ,vol. 3, pp. 31- 62 ,(1985) , 10.1145/214451.214455
E.T. Parker, A result in balanced incomplete block designs Journal of Combinatorial Theory. ,vol. 3, pp. 283- 285 ,(1967) , 10.1016/S0021-9800(67)80076-0
A.J. Smith, A Comparative Study of Set Associative Memory Mapping Algorithms and Their Use for Cache and Main Memory IEEE Transactions on Software Engineering. ,vol. SE-4, pp. 121- 130 ,(1978) , 10.1109/TSE.1978.231482
W. H. Wang, J.-L. Baer, H. M. Levy, Organization And Performance Of A Two-level Virtual-real Cache Hierarchy international symposium on computer architecture. ,vol. 17, pp. 140- 148 ,(1989) , 10.1145/74925.74942
S. McFarling, Program optimization for instruction caches architectural support for programming languages and operating systems. ,vol. 17, pp. 183- 191 ,(1989) , 10.1145/68182.68200
George Taylor, Peter Davies, Michael Farmwald, The TLB slice—a low-cost high-speed address translation mechanism ACM SIGARCH Computer Architecture News. ,vol. 18, pp. 355- 363 ,(1990) , 10.1145/325096.325161
C.R. Frink, P.J. Roy, A virtual cache-based workstation architecture [1988] Proceedings. 2nd IEEE Conference on Computer Workstations. pp. 80- 87 ,(1988) , 10.1109/COMWOR.1988.4803
Jean-Loup Baer, Wen-Hann Wang, On the inclusion properties for multi-level cache hierarchies international symposium on computer architecture. ,vol. 16, pp. 73- 80 ,(1988) , 10.1145/285930.285994