Controllable fair queuing for meeting performance goals
Published
Journal Article
Computing and storage utilities must control resource usage to meet contractual performance targets for hosted customers under dynamic conditions, including flash crowds and unexpected resource failures. This paper explores properties of proportional share resource schedulers that are necessary for stability and responsiveness under feedback control. It shows that the fairness properties commonly defined for proportional share schedulers using Weighted Fair Queuing (WFQ) are not preserved across changes to the relative weights of competing request flows. As a result, conventional WFQ schedulers are not controllable by a resource controller that adapts by adjusting the weights. The paper defines controllable fairness properties, presents an algorithm to adjust any WFQ scheduler when the weights change, and proves that the algorithm results in controllable-fair schedulers. The analytic results are confirmed by experimental evaluation using a three-tier Web service and a prototype controllable-fair scheduler called C-SFQ(D). C-SFQ(D) extends concurrency-controlled Start-time Fair Queuing (SFQ(D), which supports proportional sharing in multi-tasking computing resources. The prototype includes an adaptive control system that adjusts the flow weights in C-SFQ(D) to meet latency and throughput targets under a variety of conditions. The experimental results demonstrate the importance of controllable-fair scheduling for feedback control of computing utilities. © 2005 Elsevier B.V. All rights reserved.
Full Text
Duke Authors
Cited Authors
- Karlsson, M; Karamanolis, C; Chase, J
Published Date
- October 1, 2005
Published In
Volume / Issue
- 62 / 1-4
Start / End Page
- 278 - 294
International Standard Serial Number (ISSN)
- 0166-5316
Digital Object Identifier (DOI)
- 10.1016/j.peva.2005.07.019
Citation Source
- Scopus