Abstract
I/O consolidation is a growing trend in production environments due to increasing complexity in tuning and managing storage systems. A consequence of this trend is the need to serve multiple users and/or workloads simultaneously. It is imperative to ensure that these users are insulated from each other by virtualization in order to meet any service-level objective (SLO). Previous proposals for performance virtualization suffer from one or more of the following drawbacks: (1) They rely on a fairly detailed performance model of the underlying storage system; (2) couple rate and latency allocation in a single scheduler, making them less flexible; or (3) may not always exploit the full bandwidth offered by the storage system.This article presents a two-level scheduling framework that can be built on top of an existing storage utility. This framework uses a low-level feedback-driven request scheduler, called AVATAR, that is intended to meet the latency bounds determined by the SLO. The load imposed on AVATAR is regulated by a high-level rate controller, called SARC, to insulate the users from each other. In addition, SARC is work-conserving and tries to fairly distribute any spare bandwidth in the storage system to the different users. This framework naturally decouples rate and latency allocation. Using extensive I/O traces and a detailed storage simulator, we demonstrate that this two-level framework can simultaneously meet the latency and throughput requirements imposed by an SLO, without requiring extensive knowledge of the underlying storage system.
- Alvarez, G. A., Borowsky, E., Go, S., Romer, T. H., Becker-Szendy, R., Golding, R., Merchant, A., Spasojevic, M., Veitch, A., and Wilkes, J. 2001. Minerva: An automated resource provisioning tool for large-scale storage systems. ACM Trans. Comput. Syst. 19, 4, 483--518. Google ScholarDigital Library
- Anderson, E., Hobbs, M., Keeton, K., Spence, S., Uysal, M., and Veitch, A. 2002. Hippodrome: Running circles around storage administration. In Proceedings of the Conference on File and Storage Technology (FAST). 175--188. Google ScholarDigital Library
- Bruno, J. L., Brustoloni, J. C., Gabber, E., Ozden, B., and Silberschatz, A. 1999. Disk scheduling with quality of service guarantees. In ICMCS, Vol. 2. 400--405. Google ScholarDigital Library
- Chambliss, D., Alvarez, G., Pandey, P., Jadav, D., Xu, J., Menon, R., and Lee, T. 2003. Performance virtulization for large-scale storage systems. In Proceedings of the Symposium on Reliable Distributed Systems (SRDS).Google Scholar
- Ganger, G., Worthington, B., and Patt, Y. 2006. The DiskSim Simulation Environment Version 2.0 Reference Manual. http://www.pdl.cmu.edu/DiskSim/.Google Scholar
- Goyal, P., Jadav, D., Modha, D. S., and Tewari, R. 2003. CacheCOW: QoS for storage system caches. In Proceedings of the 8th International Workshop on Quality of Service (IWQoS), Monterey, CA.Google Scholar
- Huang, L., Peng, G., and Chiueh, T.-C. 2004. Multi-Dimensional storage virtualization. In Proceedings of the International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS). Google ScholarDigital Library
- Jin, W., Chase, J., and Kaur, J. 2004. Interposed proportional sharing for a storage service utility. In Proceedings of the International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS). Google ScholarDigital Library
- Karlsson, M., Karamanolis, C., and Zhu, X. 2004. Triage: Performance isolation and differentiation for storage systems. In Proceedings of the 9th International Workshop on Quality of Service (IWQoS).Google Scholar
- Ko, B.-J., Lee, K.-W., Amiri, K., and Calo, S. 2003. Scalable service differentiation in a shared storage cache. In Proceedings of the 23rd International Conference on Distributed Computing Systems. Google ScholarDigital Library
- Lumb, C., Merchant, A., and Alvarez, G. 2003. Facade: Virtual storage devices with performance guarantees. In Proceedings of the Conference on File and Storage Technology (FAST). 89--102. Google ScholarDigital Library
- Parekh, A. K. and Gallager, R. G. 1993. A generalized processor sharing approach to flow control in integrated services networks: The single node case. IEEE/ACM Trans. Netw., 344--357. Google ScholarDigital Library
- Shenoy, P. J. and Vin, H. M. 2002. Cello: A disk scheduling framework for next generation operating systems. Real Time Syst. J. (special issue on flexible scheduling of real-time system), 9--47. Google ScholarDigital Library
- Shriver, E., Merchant, A., and Wilkes, J. 1998. An analytical behavior model for disk drives with readahead caches and request reordering. In Proceedings of the International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS). Google ScholarDigital Library
- The Openmail Trace. 2006. http://tesla.hpl.hp.com/private_software/.Google Scholar
- Uysal, M., Alvarez, G. A., and Merchant, A. 2001. A modular, analytical throughput model for modern disk arrays. In Proceedings of the 9th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunications Systems (MASCOTS). Google ScholarDigital Library
- Wang, M., Au, K., Ailamaki, A., Brockwell, A., Faloutsos, C., and Ganger, G. 2004. Storage device performance prediction with CART models. In Proceedings of the International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS). Google ScholarDigital Library
- WebSearch trace. 2006. http://traces.cs.umass.edu/storage/.Google Scholar
- Worthington, B., Ganger, G., and Patt, Y. 1994. Scheduling algorithms for modern disk drives. In Proceedings of the International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS). Google ScholarDigital Library
- Zhang, H. 1995. Service disciplines for guaranteed performance service in packet-switching networks. In Proceedings of the IEEE 83 Conference.Google ScholarCross Ref
- Zhou, Y., Philbin, J., and Li, K. 2001. The multi-queue replacement algorithm for second level buffer caches. In Proceedings of the Usenix Technical Conference. Google ScholarDigital Library
Index Terms
- Storage performance virtualization via throughput and latency control
Recommendations
Quantifying the performance isolation properties of virtualization systems
ExpCS '07: Proceedings of the 2007 workshop on Experimental computer scienceIn this paper, we present the design of a performance isolation benchmark that quantifies the degree to which a virtualization system limits the impact of a misbehaving virtual machine on other well-behaving virtual machines running on the same physical ...
An interposed 2-Level I/O scheduling framework for performance virtualization
SIGMETRICS '05: Proceedings of the 2005 ACM SIGMETRICS international conference on Measurement and modeling of computer systemsvCache: architectural support for transparent and isolated virtual LLCs in virtualized environments
MICRO-48: Proceedings of the 48th International Symposium on MicroarchitectureA key role of virtualization is to give an illusion that a consolidated workload runs on a dedicated machine although the underlying resources are actively shared by multiple workloads. Technical advances have enabled a virtual machine (VM) to exercise ...
Comments