Abstract
Static cache partitioning can reduce inter-application cache interference and improve the composite performance of a cache-polluted application and a cache-sensitive application when they run on cores that share the last level cache in the same multi-core processor. In a virtualized system, since different applications might run on different virtual machines (VMs) in different time, it is inapplicable to partition the cache statically in advance. This paper proposes a dynamic cache partitioning scheme that makes use of hot page detection and page migration to improve the composite performance of co-hosted virtual machines dynamically according to prior knowledge of cache-sensitive applications. Experimental results show that the overhead of our page migration scheme is low, while in most cases, the composite performance is an improvement over free composition.
Similar content being viewed by others
References
Jin X, Chen H, Wang X, et al. A simple cache partitioning approach in a virtualized environment. In: Proceedings of the 2009 IEEE International Symposium on Parallel and Distributed Processing with Applications. 2009, 519–524
Bershad B, Lee D, Romer T, et al. Avoiding conflict misses dynamically in large direct-mapped caches. In: Proceedings of the 6th International Conference on Architectural Support for Programming Languages and Operating Systems. 1994, 158–170
Bugnion E, Anderson J, Mowry T, et al. Compiler-directed page coloring for multiprocessors. In: Proceedings of the 7th International Conference on Architectural Support for Programming Languages and Operating Systems. 1996, 244–255
Sherwood T, Calder B, Emer J. Reducing cache misses using hardware and software page placement. In: Proceedings of the 13th International Conference on Supercomputing. 1999, 155–164
Calder B, Krintz C, John S, et al. Cache-conscious data placement. In: Proceedings of the 8th International Conference on Architectural Support for Programming Languages and Operating Systems. 1998, 139–149
Rafique N, Lim W, Thottethodi M. Architectural support for operating system-driven CMP cache management. In: Proceedings of the 15th International Conference on Parallel Architectures and Compilation Techniques. 2006, 2–12
Barham P, Dragovic B, Fraser K, et al. Xen and the art of virtualization. In: Proceedings of the 19th ACM Symposium on Operating Systems Principles. 2003, 164–177
Jaleel A, Hasenplaugh W, Qureshi M, et al. Adaptive insertion policies for managing shared caches. In: Proceedings of the 17th International Conference on Parallel Architectures and Compilation Techniques. 2008, 208–219
Kivity A, Kamay Y, Laor D, et al. Kvm: the linux virtual machine monitor. In: Proceedings of the Linux Symposium. 2007, 225–230
Riel R. Page replacement in linux 2.4 memory management. In: Proceedings of the 2001 USENIX Annual Technical Conference — FREENIX Track. 2001, 165–172
Chen H, Wang X, Wang Z, et al. DMM: a dynamic memory mapping model for virtual machines. Science China Information Sciences, 2010, 53(5): 1097–1108
Molka D, Hackenberg D, Schone R, et al. Memory performance and cache coherency effects on an Intel Nehalem multiprocessor system. In: Proceedings of the 18th International Conference on Parallel Architectures and Compilation Techniques. 2009, 261–270
Author information
Authors and Affiliations
Corresponding author
Additional information
Xiaolin WANG received his BS and PhD degrees in Computer Science from Peking University in 1996 and 2001, respectively. Subsequently, he became a postdoctoral researcher in Peking University where he is now an associate professor. His research interests include system virtualization and cloud computing.
Xiang WEN received his BS degree in Computer Science fromPeking University in 2009. He is now an MS candidate at Peking University. His research interests focus on system virtualization and cloud computing.
Yecheng LI received his BS in Computing Science at Peking University in 2012. He is going to be a MS candidate at Peking University in Summer 2012. His research interests focus on system virtualization and cloud computing.
Zhenlin WANG is an associate professor in Michigan Technological University. He received his BS from Peking University in 1992 and his PhD from the University of Massachusetts, Amherst in 2004. His research interests focus on memory system architecture, compiler optimizations, high performance computing, and virtualization.
Yingwei LUO received his BS from Zhejiang University in 1993, and his MS and PhD in Computer Science from Peking University in 1996 and 1999, respectively. He is now a professor in Peking University. His research interests include system virtualization technologies and spatial information system.
Xiaoming LI received his BS degree from Harbin Institute of Technology in 1982 and his PhD from Stevens Institute of Technology in 1986. He is now the director of the Institute of Network Computing and Information Systems at the School of EECS, Peking University. His research interests focus on search engine technologies and peer-to-peer computing.
Rights and permissions
About this article
Cite this article
Wang, X., Wen, X., Li, Y. et al. Dynamic cache partitioning based on hot page migration. Front. Comput. Sci. 6, 363–372 (2012). https://doi.org/10.1007/s11704-012-2099-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11704-012-2099-6