EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
4024. Performance analysis of a cluster-based service system
Invited abstract in session WA-39: Stochastic Models in Service Operations II, stream Stochastic Modelling.
Wednesday, 8:30-10:00Room: 35 (building: 306)
Authors (first author is the speaker)
| 1. | Sabine Wittevrongel
|
| TELIN, Ghent University | |
| 2. | Osama Salameh
|
| Faculty of Engineering, Arab American University |
Abstract
Cluster-based service systems are used to provide parallel processing of jobs. A specific feature of such systems is that jobs are divided into tasks that should run in parallel on different servers. A job does not start execution unless the required number of idle servers is available. This paper proposes a new analytical model based on Markov chain theory that captures the typical dynamics of such cluster-based system with finite buffer size. Through our analysis, key performance measures of a cluster-based service system such as the job blocking probability, the average job delay in the queue or the average utilization of the cluster servers can be estimated. The application of the model to cloud centers with thousands of servers and a heterogeneous workload is shown possible.
Keywords
- Queuing Systems
- Stochastic Models
- Computer Science/Applications
Status: accepted
Back to the list of papers