On Identifying User Session Boundaries in Parallel Workload Logs
The stream of jobs submitted to a parallel supercomputer is actually the interleaving of many streams from different users, each of which is composed of sessions. Identifying and characterizing the sessions is important in the context of workload modeling, especially if a user-based workload model is considered. Traditionally, sessions have been delimited by long think times, that is, by intervals of more than, say, 20 minutes from the termination of one job to the submittal of the next job. We show that such a definition is problematic in this context, because jobs may be extremely long. As a result of including each job’s execution in the session, we may get unrealistically long sessions, and indeed, users most probably do not always stay connected and wait for the termination of long jobs. We therefore suggest that sessions be identified based on proven user activity, namely the submittal of new jobs, regardless of how long they run.
KeywordsSession Length Session Number Parallel Supercomputer Parallel Workload Arrival Approach
Unable to display preview. Download preview PDF.
- 2.Downey, D., Dumais, S., Horvitz, E.: Models of searching and browsing: Languages, studies, and applications. In: 20th Intl. Joint Conf. Artificial Intelligence, pp. 1465–1472 (January 2007)Google Scholar
- 6.Mehrzadi, D., Feitelson, D.G.: On extracting session data from activity logs. In: 5th Intl. Syst. & Storage Conf. (June 2012)Google Scholar
- 9.Parallel workloads archive, http://www.cs.huji.ac.il/labs/parallel/workload/
- 10.Schroeder, B., Wierman, A., Harchol-Balter, M.: Open versus closed: A cautionary tale. In: 3rd Networked Systems Design & Implementation, pp. 239–252 (May 2006)Google Scholar
- 11.Shmueli, E., Feitelson, D.G.: Using site-level modeling to evaluate the performance of parallel system schedulers. In: 14th Modeling, Anal. & Simulation of Comput. & Telecomm. Syst., pp. 167–176 (September 2006)Google Scholar
- 12.Shmueli, E., Feitelson, D.G.: Uncovering the effect of system performance on user behavior from traces of parallel systems. In: 15th Modeling, Anal. & Simulation of Comput. & Telecomm. Syst., pp. 274–280 (October 2007)Google Scholar
- 14.Shriver, E., Hansen, M.: Search Session Extraction: A User Model of Searching. Tech. rep., Bell Labs (January 2002)Google Scholar
- 16.Zilber, J., Amit, O., Talby, D.: What is worth learning from parallel workloads? a user and session based analysis. In: 19th Intl. Conf. Supercomputing, pp. 377–386 (June 2005)Google Scholar