ACM Computing Surveys (CSUR)
The working set model for program behavior
Communications of the ACM
Program behavior and virtual memory management in time-shared computer systems.
Program behavior and virtual memory management in time-shared computer systems.
Bibliography on paging and related topics
ACM SIGOPS Operating Systems Review
A Modified Locality Matrix Model (MLMM) - dynamic clustering in a demand paging environment
ACM '76 Proceedings of the 1976 annual conference
A dynamic clustering strategy in a demand paging environment
ANSS '76 Proceedings of the 4th symposium on Simulation of computer systems
ACM '77 Proceedings of the 1977 annual conference
| Hi-index | 0.00 |
The concept of information flow and information theory, adapted from the management sciences, is utilized to provide a measure of program locality that could be used to analyze, compare, and extend various replacement algorithms. A Locality Matrix Model (LMM) is derived and analyzed in the areas of page fault reductions and the determination of memory requirements to contain any inherent program locality. This approach of dynamic clustering page references is used to predict stationary behavior of the reference string.