Automated control of multiple virtualized resources

Abstract

Virtualized data centers enable sharing of resources among hosted applications. However, it is difficult to satisfy service-level objectives(SLOs) of applications on shared infrastructure, as application workloads and resource consumption patterns change over time. In this paper, we present AutoControl, a resource control system that automatically adapts to dynamic workload changes to achieve application SLOs. AutoControl is a combination of an online model estimator and a novel multi-input, multi-output (MIMO) resource controller. The model estimator captures the complex relationship between application performance and resource allocations, while the MIMO controller allocates the right amount of multiple virtualized resources to achieve application SLOs. Our experimental evaluation with RUBiS and TPC-W benchmarks along with production-trace-driven workloads indicates that AutoControl can detect and mitigate CPU and disk I/O bottlenecks that occur over time and across multiple nodes by allocating each resource accordingly. We also show that AutoControl can be used to provide service differentiation according to the application priorities during resource contention.

DOI: 10.1145/1519065.1519068

Extracted Key Phrases

16 Figures and Tables

050200920102011201220132014201520162017
Citations per Year

408 Citations

Semantic Scholar estimates that this publication has 408 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{Padala2009AutomatedCO, title={Automated control of multiple virtualized resources}, author={Pradeep Padala and Kai-Yuan Hou and Kang G. Shin and Xiaoyun Zhu and Mustafa Uysal and Zhikui Wang and Sharad Singhal and Arif Merchant}, booktitle={EuroSys}, year={2009} }