Abstract
Chandra and Toueg first investigated the problem of solving consensus in asynchronous systems with unreliable failure detectors. In practice, such failure detectors work by suspecting consensus participants that do not respond in sufficient time, i.e., within a timeout (a maximum waiting time). However, calculating this timeout is difficult. In this way, many systems explore prediction techniques to dynamically forecast the communication delay used in timeout computation. In this paper, we are concerned with the timeout associated with failure detection in asynchronous systems in which processes may crash and links may lose messages. We propose a lightweight interface to transform a non-periodic sampling of the round-trip communication delay (rtt) observed by a pull-style failure detector in periodic sampling. By using this interface, we explore the use of time series predictors to dynamically forecast the rtt. As a result, we show that these predictors are more accurate than others commonly used on failure detectors.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Fischer, M., Lynch, N., Paterson, M.: Impossibility of distributed consensus with one faulty process. Journal of the ACM 32, 374–382 (1985)
Chandra, T.D., Toueg, S.: Unreliable failure detectors for reliable distributed systems. Journal of the ACM 43, 225–267 (1996)
Raynal, M., Tronel, F.: Group membership failure detecton: a simple protocol and its probabilistic analysis. Distributed Systems Engineering Journal 6, 95–102 (1999)
van Renesse, R., Minsky, Y., Hayden, M.: A gossip-style failure detection service. In: Proc. of the Middleware Conference (1998), Available at Cornell Digital Library (TR98-1687)
Macêdo, R.: Implementing failure detection through the use of a self-tuned time connectivity indicator. Technical Report RT008/98, Laboratório de Sistemas Distribuídos – LaSiD, Salvador-Brazil (1998)
Chen, W., Toueg, S., Aguilera, M.K.: On the quality of service of failure detectors. IEEE Transaction on Computers 51, 561–580 (2002)
Jacobson, V.: Congestion avoidance and control. In: Proc. of the ACM Symposium on Communications, Architectures and Protocols (ACM SIGCOMM), Palo Alto, pp. 314–329. ACM Press, New York (1988); Slightly-revised 1992 version of the 1988 paper
Felber, P., Défago, X., Guerraoui, R., Oser, P.: Failure detectors as first class objects. In: Proc. of the Intl. Symp. on Distributed Objects and Applications – DOA 1999. IEEE Computer Society Press, Los Alamitos (1999)
Nunes, R.C., Jansch-Porto, I.: Modelling communication delays in distributed systems using time series. In: Proceedings of the 21th IEEE Symposium on Reliable Distributed Systems (SRDS 2002), Osaka-Japan, pp. 268–273 (2002)
Bowerman, B.L., O’Connel, R.T.: Forecasting and Time Series: an Applied Approach, 3rd edn. Duxbury Press, Belmont (1993)
Dinda, P.A., O’Hallaron, D.R.: An extensible toolkit for resource prediction in distributed systems. Technical Report MU-CS-99-13, Computer Science Department – Carnegie Mellon University, Pittsburgh (1999)
Schulz, J., Hochberger, C., Tavangarian, D.: Prediction of communication performance for wide area computing systems. In: Proceedings of the 9th Euromicro Workshop on Parallel and Distributed Processing, Mantova-Italy, pp. 480–486. IEEE Computer Society Press, Los Alamitos (2001)
Bauer, P., Sichitiu, M., Premaratne, K.: On the nature of the time-variant communication delays. In: Proceedings of the IASTED Conference Modeling, Identifi cation and Control (MIC 2001), Innsbruck, Austria, pp. 792–797 (2001)
Ryan, K.F., Giles, D.E.A.: Testing for unit roots with missing observations. In: Fomby, T.B., Carter Hill, R. (eds.) Advances in Econometrics (1998) (forthcoming)
Nunes, R.C., Jansch-Porto, I.: Non-stationary communication delays in failure detectors. In: Proceedings of the 3rd IEEE Latin-American test Workshop (LATW 2002), Montevideo-Uruguay, pp. 16–21 (2002)
Montgomery, D.C., Johnson, L.A.: Forecasting and time series analysis. McGraw-Hill, New York (1976)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nunes, R.C., Jansch-Pôrto, I. (2003). A Lightweight Interface to Predict Communication Delays Using Time Series. In: de Lemos, R., Weber, T.S., Camargo, J.B. (eds) Dependable Computing. LADC 2003. Lecture Notes in Computer Science, vol 2847. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-45214-0_19
Download citation
DOI: https://doi.org/10.1007/978-3-540-45214-0_19
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-20224-0
Online ISBN: 978-3-540-45214-0
eBook Packages: Springer Book Archive