Print |  Close this window

Adaptive Control of Virtualized Resources in Utility Computing Environments

Authors (if Research)
Pradeep Padala, Xiaoyun Zhu, et al.
Operating Regions
Operating Countries
Proceedings of the EuroSys 2007. ABSTRACT: Data centers are often under-utilized due to over-provisioning as well as time-varying resource demands of typical enter- prise applications. One approach to increase resource uti- lization is to consolidate applications in a shared infrastruc- ture using virtualization. Meeting application-level quality of service (QoS) goals becomes a challenge in a consolidated environment as application resource needs differ. Further- more, for multi-tier applications, the amount of resources needed to achieve their QoS goals might be different at each tier and may also depend on availability of resources in other tiers. In this paper, we develop an adaptive resource con- trol system that dynamically adjusts the resource shares to individual tiers in order to meet application-level QoS goals while achieving high resource utilization in the data cen- ter. Our control system is developed using classical control theory, and we used a black-box system modeling approach to overcome the absence of first principle models for com- plex enterprise applications and systems. To evaluate our controllers, we built a testbed simulating a virtual data cen- ter using Xen virtual machines. We experimented with two multi-tier applications in this virtual data center: a two- tier implementation of RUBiS, an online auction site, and a two-tier Java implementation of TPC-W. Our results in- dicate that the proposed control system is able to maintain high resource utilization and meets QoS goals in spite of varying resource demands from the applications.
Print |  Close this window