A case for integrated processor-cache partitioning in chip multiprocessors
Authors:
Open Access
- 14 November 2009
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-12
Abstract
Existing cache partitioning schemes are designed in a manner oblivious to the implicit processor partitioning enforced by the operating system. This paper examines an operating system directed integrated processor-cache partitioning scheme that partitions both the available processors and the shared cache in a chip multiprocessor among different multi-threaded applications. Extensive simulations using a set of multiprogrammed workloads show that our integrated processor-cache partitioning scheme facilitates achieving better performance isolation as compared to state of the art hardware/software based solutions. Specifically, our integrated processor-cache partitioning approach performs, on an average, 20.83% and 14.14% better than equal partitioning and the implicit partitioning enforced by the underlying operating system, respectively, on the fair speedup metric on an 8 core system. We also compare our approach to processor partitioning alone and a state-of-the-art cache partitioning scheme and our scheme fares 8.21% and 9.19% better than these schemes on a 16 core system.Keywords
Funding Information
- Division of Computer and Network Systems (CNS 0720645CCF 0811687CCF 0702519CNS 0202007CNS 0509251)
- Division of Computing and Communication Foundations (CNS 0720645CCF 0811687CCF 0702519CNS 0202007CNS 0509251)
This publication has 28 references indexed in Scilit:
- Fairness enforcement in switch on event multithreadingACM Transactions on Architecture and Code Optimization, 2007
- QoS policies and architecture for cache/memory in CMP platformsACM SIGMETRICS Performance Evaluation Review, 2007
- Adaptive control of virtualized resources in utility computing environmentsACM SIGOPS Operating Systems Review, 2007
- From chaos to QoSACM SIGARCH Computer Architecture News, 2007
- The Impact of Performance Asymmetry in Emerging Multicore ArchitecturesACM SIGARCH Computer Architecture News, 2005
- Dynamic Partitioning of Shared Cache MemoryThe Journal of Supercomputing, 2004
- More on finding a single number to indicate overall performance of a benchmark suiteACM SIGARCH Computer Architecture News, 2004
- Simics: A full system simulation platformComputer, 2002
- Characterizing computer performance with a single numberCommunications of the ACM, 1988
- A fair share schedulerCommunications of the ACM, 1988