Abstract
The restart method is based purely on the task completion time. In the previous chapter conditions under which the method is triggered have been investigated. When applying the restart method the only relevant question from a quantitative analysis point is when to restart. For the task under consideration its processing time is monitored and if the processing time exceeds a given value τ then the task is aborted and restarted from beginning. In stochastic terms one may say restart uses a completion time distribution, and a job is restarted when the remaining expected completion time is longer than the expected completion time when restarting the job, taking into account the time already waited for completion. This is similar to age replacement in preventive maintenance.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
At the cost of more notation, and with a proper discussion for special cases, the results in this section also apply to distributions defined over finite domains, as well as to defective distributions and distributions with jumps.
- 2.
At times we also refer to τ as the as restartinterval.
- 3.
There is no particular significance to the chosen parameter values. They happen to be the parameters of a lognormal fit for experimental data of HTTP GET completion times [128].
- 4.
Note that because of the workings of the Mathematica optimisation algorithm, the comparison in Fig.4.14 had to be based on convergence of\({\rm E}[{T_K^N}]\) as stopping criterion, while we were able to base the results for the backward/forward algorithm in Fig.4.15 on convergence of restart times, a stricter criterion. This explains the higher CPU usage for the backward algorithm in Fig.4.15 compared to Fig.4.14.
- 5.
Adhering to SOAP terminology, we do not call HTTP an application, but rather a transport for the layers above it.
- 6.
Where\({\rm argmin}_{i} \left\{x_{i\in \mathbb{I}}\right\} := \min \mathbb{I}_{\mathrm{Min}}\) with
$$\begin{aligned}\mathbb{I}_{\mathrm{Min}} :=& \left\{i\in \mathbb{I} : x_i \in Min\right\}\\ Min :=& \min\left\{x_i : i\in \mathbb{I}\right\}.\end{aligned}$$
Reference
N. Looker, J. Xiu,Assessing the Dependability of Soap-Rpc-Based Web Services by Fault Injection. InProceedings of the 9th IEEE International Workshop on Object-oriented Real-time Dependable Systems, Anacapri, Italy (IEEE Computer Society, Washington, DC, 2003), pp. 163–170
B. Krishnamurthy, J. Rexford,Web Protocols and Practice (Addison-Wesley, Reading, MA, 2001)
I. Gertsbakh,Reliability Theory, with Applications to Preventive Maintenance (Springer, Berlin, 2000)
P. Reinecke, A.P.A. van Moorsel, K. Wolter,Experimental Analysis of the Correlation of HTTP GET Invocations, ed. by M. Telek, A. Horvath. InProceedings of the 3rd European Performance Engineering Workshop, Budapest, Hungary. Lecture Notes in Computer Science, vol. 4054 (Springer, Heidelberg, 2006)
BEA Systems, IBM, Microsoft Corporation Inc, TIBCO Software Inc. Web Services Reliable Messaging Protocol (WS-Reliable Messaging), February 2005
A.P.A. van Moorsel, K. Wolter,Analysis and Algorithms for Restart. InQEST’04: Proceedings of the 1st International Conference on the Quantitative Evaluation of Systems, Twente, The Netherlands, September 2004 (IEEE Computer Society, Los Alamitos, CA, 2004), pp. 195–204. (Best paper award)
W. Chen, S. Toueg, M.K. Aguilera,On the Quality of Service of Failure Detectors. InDSN’00: Proceedings of the International Conference on Dependable Systems and Networks, New York (IEEE Computer Society, Los Alamitos, CA, 2000), pp. 191–200
V. Paxson,End-to-End Internet Packet Dynamics. InProceedings of the ACM SIGCOMM’97 Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication, Cannes, France, September 1997 (ACM Press, New York, 1997); Comput. Commun. Rev.27(4), 139–154
V.F. Nicola, Chapter 7: Checkpointing and the modeling of program execution time, inSoftware Fault Tolerance, ed. by M.R. Lyu.Trends in Software, vol. 3 (Wiley, Chichester, 1995), pp. 167–188
M. Allman, V. Paxson,On Estimating End-to-End Network Path Properties. InSIGCOMM’99: Proceedings of the Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication, New York, September 1999 (ACM Press, New York, 1999), pp. 263–274
V. Paxson,On Calibrating Measurements of Packet Transit Times. InSIGMETRICS’98/PERFORMANCE ’98: Proceedings of the ACM SIGMETRICS Joint International Conference on Measurement and Modeling of Computer Systems, Madison, WI (ACM Press, New York, 1998), pp. 11–21
B.R. Haverkort,Performance of Computer Communication Systems: A Model-Based Approach (Wiley, Chichester, 1998)
P. Reinecke, A. van Moorsel, K. Wolter,A Measurement Study of the Interplay Between Application Level Restart and Transport Protocol. InISAS’04: Proceedings of the International Service Availability Symposium, Munich, Germany, May 2004. Lecture Notes in Computer Science, vol. 3335 (Springer, Heidelberg, 2004)
The Apache Software Foundation: Apache Sandesha,http://ws.apache.org/sandesha/
The Apache Software Foundation: Apache Axis,http://ws.apache.org/axis/
M. Abramowitz, I.A. Stegun (eds.),Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th edn. (Dover, New York, 1972)
A.M. Law, W.D. Kelton,Simulation Modeling & Analysis, 2nd edn. Industrial Engineering and Management Science (McGraw-Hill, New York, 1991)
S.M. Maurer, B.A. Huberman, Restart strategies and internet congestion. J. Econ. Dyn. Control25, 641–654 (2001)
Y. Ruan, E. Horvitz, H. Kautz,Restart Policies with Dependence Among Runs: A Dynamic Programming Approach. InProceedings of the 8th International Conference on Principles and Practice of Constraint Programming, Ithaca, NY, September 2002. Lecture Notes in Computer Science, vol. 2470 (Springer, Heidelberg, 2002), pp. 573–586
P. Karn, C. Partridge, Improving round-trip time estimates in reliable transport protocols.ACM Trans. Comput. Syst.9(4), 364–373 (1991)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Wolter, K. (2010). Moments of Completion Time Under Restart. In: Stochastic Models for Fault Tolerance. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-11257-7_4
Download citation
DOI: https://doi.org/10.1007/978-3-642-11257-7_4
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-11256-0
Online ISBN: 978-3-642-11257-7
eBook Packages: Computer ScienceComputer Science (R0)