Abstract
We propose a tunable scheduling strategy that lies between FIFO and shortest-first, based on the value of a coefficiert Alpha. If Alpha is set to zero then this strategy is just FIFO. Larger Alpha gets us closer to shortest-first strategy which is known to provide optimal response time. However, unlike the shortest-first, proposed scheduling strategy is starvation free. This scheduling strategy, called Alpha scheduling with no preemption, allows to improve overall response time per HTTP request more than 3 times under heavy load. We demonstrate our results with a simple simulation model using SpecWeb96 to generate representative WWW workload.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Cherkasova, L. and Rokicki, T.: Alpha Message Scheduling for Packet-Switched Interconnects. HP Laboratories Report No. HPL-94-71, August, 1994.
Fielding, R., Gettys, J., Mogul, J., Frystyk, H., Berners-Lee, T.: Hypertext Transfer Protocol-HTTP/1.1. Internet Proposed Standard Protocol, RFC-2068. URL http://www.ics.uci.edu/pub/ietf/http/.
Prefect, F., Doan, L., Gold, S., Wicki, Th., Wilcke, W.: Performance Limiting Factors in HTTP (Web)Server Operations. In Proceedingds of COMPCON96, Santa Clara, California, February, 1996, p.267–272.
TheWorkloadfortheSPECweb96Benchmark.URL http://www.sepcbench.org/osg/web96/workload.html.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Cherkasova, L. (1998). Scheduling strategy to improve response time for web applications. In: Sloot, P., Bubak, M., Hertzberger, B. (eds) High-Performance Computing and Networking. HPCN-Europe 1998. Lecture Notes in Computer Science, vol 1401. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0037157
Download citation
DOI: https://doi.org/10.1007/BFb0037157
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64443-9
Online ISBN: 978-3-540-69783-1
eBook Packages: Springer Book Archive