Trace-driven memory simulation: a survey
ACM Computing Surveys (CSUR)
Adaptive page replacement based on memory reference behavior
SIGMETRICS '97 Proceedings of the 1997 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
EELRU: simple and effective adaptive page replacement
SIGMETRICS '99 Proceedings of the 1999 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Sequentiality and prefetching in database systems
ACM Transactions on Database Systems (TODS)
Proceedings of the 2000 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
The working set model for program behavior
Communications of the ACM
SIGMETRICS '02 Proceedings of the 2002 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
IEEE Transactions on Computers
2Q: A Low Overhead High Performance Buffer Management Replacement Algorithm
VLDB '94 Proceedings of the 20th International Conference on Very Large Data Bases
General adaptive replacement policies
Proceedings of the 4th international symposium on Memory management
ARC: A Self-Tuning, Low Overhead Replacement Cache
FAST '03 Proceedings of the 2nd USENIX Conference on File and Storage Technologies
CAR: Clock with Adaptive Replacement
FAST '04 Proceedings of the 3rd USENIX Conference on File and Storage Technologies
CLOCK-Pro: an effective improvement of the CLOCK replacement
ATEC '05 Proceedings of the annual conference on USENIX Annual Technical Conference
OSDI'00 Proceedings of the 4th conference on Symposium on Operating System Design & Implementation - Volume 4
An implementation study of a detection-based adaptive block replacement scheme
ATEC '99 Proceedings of the annual conference on USENIX Annual Technical Conference
Operating System Concepts
Hi-index | 0.00 |
Adaptive algorithms are capable of modifying their own behavior through time, depending on the execution characteristics. Recently, we have proposed LRU-WAR, an adaptive replacement algorithm whose objective is to minimize failures detected in LRU policy, preserving its simplicity and low overhead. In this paper, we present our contribution to the study of adaptive replacement algorithms describing their behavior under a number of workloads. Simulations include an analysis of the performance sensibility with the variation of the control parameters and its application in a multiprogrammed environment. In order to address LRU-WAR weakness as a global policy, we also introduce LRU-WARlock. The simulation results show that substantial performance improvements can be obtained.