Improved Prediction for Web Server Delay Control
(2004) Proceedings. 16th Euromicro Conference on Real-Time Systems p.61-68- Abstract
- Control methods are being used increasingly for uncertainty management and QoS in modern web server systems. Previous approaches have suggested combined feed-forward and feedback control strategies, using queuing theory for feed-forward delay prediction. While queuing theory allows one to predict delay as a function of arrival and service rates, the prediction applies only to long-term averages, and is therefore insensitive to sudden load changes. Unfortunately, Internet load is very bursty, leaving room for predictor improvement.The main contribution of this paper is an extension of the combined feed-forward/feedback framework in which the queuing model is replaced with a predictor that instead uses instantaneous measurements to predict... (More)
- Control methods are being used increasingly for uncertainty management and QoS in modern web server systems. Previous approaches have suggested combined feed-forward and feedback control strategies, using queuing theory for feed-forward delay prediction. While queuing theory allows one to predict delay as a function of arrival and service rates, the prediction applies only to long-term averages, and is therefore insensitive to sudden load changes. Unfortunately, Internet load is very bursty, leaving room for predictor improvement.The main contribution of this paper is an extension of the combined feed-forward/feedback framework in which the queuing model is replaced with a predictor that instead uses instantaneous measurements to predict future delays. The proposed strategy is evaluated in simulation and by experiments on an Apache web server. It is shown that the new approach performs better than the combined queuing model based feed-forward and feedback control presented in earlier papers. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/535953
- author
- Henriksson, Dan LU ; Lu, Ying and Abdelzaher, Tarek
- organization
- publishing date
- 2004
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- delay prediction, queuing theory, Internet, feedback control, feedforward control, QoS, uncertainty management, Web server, delay control
- host publication
- Proceedings of the 16th Euromicro Conference on Real-Time Systems (ECRTS 04)
- pages
- 61 - 68
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- Proceedings. 16th Euromicro Conference on Real-Time Systems
- conference location
- Catania, Italy
- conference dates
- 2004-06-30 - 2004-07-02
- external identifiers
-
- wos:000222970300007
- scopus:5644249618
- ISSN
- 1068-3070
- ISBN
- 0-7695-2176-2
- DOI
- 10.1109/EMRTS.2004.1311001
- language
- English
- LU publication?
- yes
- id
- 469ff1a0-dcd6-4c6c-8d9e-d85193646c42 (old id 535953)
- alternative location
- http://citeseer.ist.psu.edu/henriksson04improved.html
- date added to LUP
- 2016-04-01 15:29:42
- date last changed
- 2022-01-28 05:38:09
@inproceedings{469ff1a0-dcd6-4c6c-8d9e-d85193646c42, abstract = {{Control methods are being used increasingly for uncertainty management and QoS in modern web server systems. Previous approaches have suggested combined feed-forward and feedback control strategies, using queuing theory for feed-forward delay prediction. While queuing theory allows one to predict delay as a function of arrival and service rates, the prediction applies only to long-term averages, and is therefore insensitive to sudden load changes. Unfortunately, Internet load is very bursty, leaving room for predictor improvement.The main contribution of this paper is an extension of the combined feed-forward/feedback framework in which the queuing model is replaced with a predictor that instead uses instantaneous measurements to predict future delays. The proposed strategy is evaluated in simulation and by experiments on an Apache web server. It is shown that the new approach performs better than the combined queuing model based feed-forward and feedback control presented in earlier papers.}}, author = {{Henriksson, Dan and Lu, Ying and Abdelzaher, Tarek}}, booktitle = {{Proceedings of the 16th Euromicro Conference on Real-Time Systems (ECRTS 04)}}, isbn = {{0-7695-2176-2}}, issn = {{1068-3070}}, keywords = {{delay prediction; queuing theory; Internet; feedback control; feedforward control; QoS; uncertainty management; Web server; delay control}}, language = {{eng}}, pages = {{61--68}}, publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}}, title = {{Improved Prediction for Web Server Delay Control}}, url = {{https://lup.lub.lu.se/search/files/4405301/625603.pdf}}, doi = {{10.1109/EMRTS.2004.1311001}}, year = {{2004}}, }