Abstract
Physical memory is the scarcest resource in today's cloud computing platforms. Cloud providers would like to maximize their clients' satisfaction by renting precious physical memory to those clients who value it the most. But real-world cloud clients are selfish: they will only tell their providers the truth about how much they value memory when it is in their own best interest to do so. How can real-world cloud providers allocate memory efficiently to those (selfish) clients who value it the most?
We present Ginseng, the first market-driven cloud system that allocates memory efficiently to selfish cloud clients. Ginseng incentivizes selfish clients to bid their true value for the memory they need when they need it. Ginseng continuously collects client bids, finds an efficient memory allocation, and re-allocates physical memory to the clients that value it the most. Ginseng achieves a 6.2×--15.8x improvement (83%--100% of the optimum) in aggregate client satisfaction when compared with state-of-the-art approaches for cloud memory allocation.
- O. Agmon Ben-Yehuda, M. Ben-Yehuda, A. Schuster, and D. Tsafrir. Deconstructing Amazon EC2 spot instance pricing. In IEEE Conf. on Cloud Computing Technology and Science (CloudCom), 2011. Google Scholar
Digital Library
- O. Agmon Ben-Yehuda, M. Ben-Yehuda, A. Schuster, and D. Tsafrir. The resource-as-a-service (raas) cloud. In USENIX Conf. on Hot Topics in Cloud Computing (HotCloud), 2012. Google Scholar
Digital Library
- M. Armbrust, A. Fox, R. Griffith, A. D. Joseph, R. Katz, A. Konwinski, G. Lee, D. Patterson, A. Rabkin, I. Stoica, and M. Zaharia. A view of cloud computing. Communications of the ACM, 53 (4): 50--58, 2010. Google Scholar
Digital Library
- J. Bae, E. Beigman, R. Berry, M. L. Honig, and R. Vohra. An efficient auction for non concave valuations. In Int'l Meeting of the Society for Social Choice and Welfare, 2008.Google Scholar
- M. Cary, A. Das, B. Edelman, I. Giotis, K. Heimerl, A. R. Karlin, C. Mathieu, and M. Schwarz. Greedy bidding strategies for keyword auctions. In ACM Conf. on Electronic Commerce (EC), pages 262--271, 2007. Google Scholar
Digital Library
- J. S. Chase, D. C. Anderson, P. N. Thakar, A. M. Vahdat, and R. P. Doyle. Managing energy and server resources in hosting centers. In ACM Symposium on Operating Systems Principles (SOSP), 2001. Google Scholar
Digital Library
- E. H. Clarke. Multipart pricing of public goods. Public Choice, 11 (1): 17--33, Sep 1971.Google Scholar
Cross Ref
- G. D'Alesandre. Updated app engine pricing faq! Web site, June 2011. http://tinyurl.com/D-Alesandre.Google Scholar
- S. Dobzinski and N. Nisan. Mechanisms for multi-unit auctions. J. of Artificial Intelligence Research, 37: 85--98, 2010. Google Scholar
Digital Library
- K. E. Drexler and M. S. Miller. Incentive engineering for computational resource management. In The Ecology of Computation, pages 231--266. Elsevier Science Publishers, 1988.Google Scholar
- Z. Gong, X. Gu, and J. Wilkes. Press: Predictive elastic resource scaling for cloud systems. In Int'l Conf. on Network and Service Management (CNSM), pages 9--16, 2010.Google Scholar
- A. Gordon, M. Hines, D. Da Silva, M. Ben-Yehuda, M. Silva, and G. Lizarraga. Ginkgo: Automated, application-driven memory overcommitment for cloud computing. In Runtime Environments/Systems, Layering, & Virtualized Environments Workshop (ASPLOS RESOLVE), 2011.Google Scholar
- T. Groves. Incentives in teams. Econometrica, 41 (4): 617--631, Jul 1973.Google Scholar
Cross Ref
- C. Grzegorczyk, S. Soman, C. Krintz, and R. Wolski. Isla vista heap sizing: Using feedback to avoid paging. In Int'l Symposium on Code Generation and Optimization (CGO), pages 325--340, 2007. Google Scholar
Digital Library
- J. Heo, X. Zhu, P. Padala, and Z. Wang. Memory overbooking and dynamic control of xen virtual machines in consolidated environments. In IFIP/IEEE Symposium on Integrated Management (IM), 2009. Google Scholar
Digital Library
- M. Hertz, S. Kane, E. Keudel, T. Bai, C. Ding, X. Gu, and J. E. Bard. Waste not, want not: resource-based garbage collection in a shared environment. In Int'l Symposium on Memory Management (ISMM), 2011. Google Scholar
Digital Library
- M. Hines, A. Gordon, M. Silva, D. D. Silva, K. D. Ryu, and M. Ben-Yehuda. Applications know best: Performance-driven memory overcommit with ginkgo. In IEEE Conf. on Cloud Computing Technology and Science (CloudCom), 2011. Google Scholar
Digital Library
- S. T. Jones, A. C. Arpaci-Dusseau, and R. H. Arpaci-Dusseau. Geiger: monitoring the buffer cache in a virtual machine environment. In ACM Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2006. Google Scholar
Digital Library
- F. Kelly. Charging and rate control for elastic traffic. European Trans. on Telecommunications, 8: 33--37, 1997.Google Scholar
Cross Ref
- A. Kivity, Y. Kamay, D. Laor, U. Lublin, and A. Liguori. KVM: the Linux virtual machine monitor. In Ottawa Linux Symposium (OLS), pages 225--230, 2007.Google Scholar
- A. Lazar and N. Semret. Design and analysis of the progressive second price auction for network bandwidth sharing. Telecommunication Systems--Special issue on Network Economics, 20: 255--263, 1999.Google Scholar
- M. Levy and S. Solomon. New evidence for the power-law distribution of wealth. Physica A, 242: 90--94, 1997.Google Scholar
Cross Ref
- A. G. Litke. Memory overcommitment manager. website, 2011. https://github.com/aglitke/mom.Google Scholar
- B. Lucier, R. Paes Leme, and E. Tardos. On revenue in the generalized second price auction. In Int'l World Wide Web Conf. (WWW), 2012. Google Scholar
Digital Library
- D. Magenheimer. Memory overcommit... without the commitment. In Xen Summit. USENIX association, June 2008.Google Scholar
- P. Maillé and B. Tuffin. Multi-bid auctions for bandwidth allocation in communication networks. In IEEE INFOCOM, 2004.Google Scholar
Cross Ref
- R. Nathuji, A. Kansal, and A. Ghaffarkhah. Q-clouds: Managing performance interference effects for qos-aware clouds. In ACM SIGOPS European Conf. on Computer Systems (EuroSys), 2010. Google Scholar
Digital Library
- Z. Ou, H. Zhuang, J. K. Nurminen, A. Ylä-Jääski, and P. Hui. Exploiting hardware heterogeneity within the same instance type of amazon EC2. In USENIX Conf. on Hot Topics in Cloud Computing (HotCloud), 2012. Google Scholar
Digital Library
- T.-I. Salomie, G. Alonso, T. Roscoe, and K. Elphinstone. Application level ballooning for efficient server consolidation. In ACM SIGOPS European Conf. on Computer Systems (EuroSys), pages 337--350, 2013. Google Scholar
Digital Library
- Z. Shen, S. Subbiah, X. Gu, and J. Wilkes. Cloudscale: elastic resource scaling for multi-tenant cloud systems. In ACM Symposium on Cloud Computing (SOCC), 2011. Google Scholar
Digital Library
- S. Soman, C. Krintz, and D. F. Bacon. Dynamic selection of application-specific garbage collectors. In Int'l Symposium on Memory Management (ISMM), 2004. Google Scholar
Digital Library
- W. Souma. Universal structure of the personal income distribution. Fractals, 9 (04): 463--470, 2001.Google Scholar
Cross Ref
- B. Urgaonkar, P. Shenoy, and T. Roscoe. Resource overbooking and application profiling in a shared internet hosting platform. ACM Trans. Internet Technol., 9 (1), 2009. Google Scholar
Digital Library
- W. Vickrey. Counterspeculation, auctions, and competitive sealed tenders. J. of Finance, 16 (1), 1961.Google Scholar
Cross Ref
- C. Waldspurger, T. Hogg, B. A. Huberman, J. O. Kephart, and W. S. Stornetta. Spawn: a distributed computational economy. IEEE Trans. on Software Engineering, 18 (2): 103--117, 1992. Google Scholar
Digital Library
- C. A. Waldspurger. Lottery and Stride Scheduling: Flexible Proportional-Share Resource Management. PhD thesis, Massachusetts Institute of Technology, 1995. Google Scholar
Digital Library
- C. A. Waldspurger. Memory resource management in Vmware ESX server. In USENIX Symposium on Operating Systems Design & Implementation (OSDI), 2002. Google Scholar
Digital Library
- T. Yang, E. D. Berger, S. F. Kaplan, and J. E. B. Moss. CRAMM: virtual memory support for garbage-collected applications. In USENIX Symposium on Operating Systems Design & Implementation (OSDI), pages 103--116, 2006. Google Scholar
Digital Library
- W. Zhao and Z. Wang. Dynamic memory balancing for virtual machines. In ACM/USENIX Int'l Conf. on Virtual Execution Environments (VEE), pages 21--30, 2009. Google Scholar
Digital Library
- W. Zhao, X. Jin, Z. Wang, X. Wang, Y. Luo, and X. Li. Low cost working set size tracking. In USENIX Annual Technical Conf. (ATC), 2011. Google Scholar
Digital Library
- P. Zhou, V. Pandey, J. Sundaresan, A. Raghuraman, Y. Zhou, and S. Kumar. Dynamic tracking of page miss ratio curve for memory management. In ACM Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2004. Google Scholar
Digital Library
Index Terms
Ginseng: market-driven memory allocation
Recommendations
Ginseng: market-driven memory allocation
VEE '14: Proceedings of the 10th ACM SIGPLAN/SIGOPS international conference on Virtual execution environmentsPhysical memory is the scarcest resource in today's cloud computing platforms. Cloud providers would like to maximize their clients' satisfaction by renting precious physical memory to those clients who value it the most. But real-world cloud clients ...
Ginseng: market-driven LLC allocation
USENIX ATC '16: Proceedings of the 2016 USENIX Conference on Usenix Annual Technical ConferenceCloud providers must dynamically allocate their physical resources to the right client to maximize the benefit that they can get out of given hardware. Cache Allocation Technology (CAT) makes it possible for the provider to allocate last level cache to ...
DMSS: A Dynamic Memory Scheduling System in Server Consolidation Environments
ISORCW '12: Proceedings of the 2012 IEEE 15th International Symposium on Object/Component/Service-Oriented Real-Time Distributed Computing WorkshopsAs the foundation of cloud computing, Server consolidation allows multiple computer infrastructures running as virtual machines in a single physical node. It improves the utilization of most kinds of resource but memory under x86 architecture. Because ...







Comments