Fully decentralized emulation of best-effort and processor sharing queues

Stanojević, Rade and Shorten, Robert N. (2008) Fully decentralized emulation of best-effort and processor sharing queues. In: Proceedings of the 2008 ACM SIGMETRICS international conference on Measurement and modeling of computer systems. ACM, New York. ISBN 978-1-60558-005-0

[img] Download (670kB)

Share your research

Twitter Facebook LinkedIn GooglePlus Email more...

Add this article to your Mendeley library


Control of large distributed cloud-based services is a chal- lenging problem. The Distributed Rate Limiting (DRL) paradigm was recently proposed as a mechanism for tack- ling this problem. The heuristic nature of existing DRL solutions makes their behavior unpredictable and analyti- cally untractable. In this paper we treat the DRL prob- lem in a mathematical framework and propose two novel DRL algorithms that exhibit good and predictable perfor- mance. The ¯rst algorithm Cloud Control with Constant Probabilities (C3P) solves the DRL problem in best e®ort environments, emulating the behavior of a single best-e®ort queue in a fully distributed manner. The second problem we approach is the DRL in processor sharing environments. Our algorithm, Distributed Deficit Round Robin (D2R2), parameterized by parameter ®, converges to a state that is, at most, O( 1 ®) away from the exact emulation of central- ized processor sharing queue. The convergence and stability properties are fully analyzed for both C3P and D2R2. An- alytical results are validated empirically through a number of representative packet level simulations. The closed-form nature of our results allows simple design rules which, to- gether with extremely low communication overhead, makes the presented algorithms practical and easy to deploy.

Item Type: Book Section
Keywords: Rate limiting; CDN; Cloud control; Consensus agreement; Stability and convergence; Hamilton Institute.
Academic Unit: Faculty of Science and Engineering > Research Institutes > Hamilton Institute
Faculty of Science and Engineering > Mathematics and Statistics
Item ID: 1732
Depositing User: Hamilton Editor
Date Deposited: 10 Dec 2009 15:10
Publisher: ACM
Refereed: Yes

    Repository Staff Only(login required)

    View Item Item control page

    Document Downloads

    More statistics for this item...