Evaluation of NUMA memory management through modeling and measurements
- 1 January 1992
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 3 (6), 686-701
- https://doi.org/10.1109/71.180624
Abstract
Dynamic page placement policies for NUMA (nonuniform memory access time)shared-memory architectures are explored using two approaches that complement eachother in important ways. The authors measure the performance of parallel programsrunning on the experimental DUnX operating system kernel for the BBN GP1000, whichsupports a highly parameterized dynamic page placement policy. They also develop andapply an analytic model of memory system performance of a local/remote NUMAarchitecture based on approximate mean-value analysis techniques. The model isvalidated against experimental data obtained with DUnX while running a syntheticworkload. The results of this validation show that, in general, model predictions are quitegood. Experiments investigating the effectiveness of dynamic page-placement and, inparticular, dynamic multiple-copy page placement the cost of replication/coherency faulterrors, and the cost of errors in deciding whether a page should move or be remotelyreferenced are described.Keywords
This publication has 20 references indexed in Scilit:
- Memory consistency and event ordering in scalable shared-memory multiprocessorsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- An analysis of dynamic page placement on a NUMA multiprocessorPublished by Association for Computing Machinery (ACM) ,1992
- Experimental comparison of memory management policies for NUMA multiprocessorsACM Transactions on Computer Systems, 1991
- Performance evaluation of memory consistency models for shared-memory multiprocessorsPublished by Association for Computing Machinery (ACM) ,1991
- Memory access dependencies in shared-memory multiprocessorsIEEE Transactions on Software Engineering, 1990
- Analysis of critical architectural and programming parameters in a hierarchicalPublished by Association for Computing Machinery (ACM) ,1990
- The implementation of a coherent memory abstraction on a NUMA multiprocessor: experiences with platinumPublished by Association for Computing Machinery (ACM) ,1989
- Reference history, page size, and migration daemons in local/remote architecturesPublished by Association for Computing Machinery (ACM) ,1989
- Page table management in local/remote architecturesPublished by Association for Computing Machinery (ACM) ,1988
- An accurate and efficient performance analysis technique for multiprocessor snooping cache-consistency protocolsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1988