Multi-step ahead response time prediction for single server queuing systems
(2011) The 16th IEEE Symposium on Computers and Communications (ISCC)- Abstract
- Multi-step ahead response time prediction of CPU
constrained computing systems is vital for admission control,
overload protection and optimization of resource allocation in
these systems. CPU constrained computing systems such as web
servers can be modeled as single server queuing systems. These
systems are stochastic and nonlinear. Thus, a well-designed nonlinear
prediction scheme would be able to represent the dynamics
of such a system much better than a linear scheme. A nonlinear
autoregressive neural network with exogenous inputs based
multi-step ahead response time predictor has been developed.
The proposed estimator has many promising... (More) - Multi-step ahead response time prediction of CPU
constrained computing systems is vital for admission control,
overload protection and optimization of resource allocation in
these systems. CPU constrained computing systems such as web
servers can be modeled as single server queuing systems. These
systems are stochastic and nonlinear. Thus, a well-designed nonlinear
prediction scheme would be able to represent the dynamics
of such a system much better than a linear scheme. A nonlinear
autoregressive neural network with exogenous inputs based
multi-step ahead response time predictor has been developed.
The proposed estimator has many promising characteristics that
make it a viable candidate for being implemented in admission
control products for computing systems. It has a simple structure,
is nonlinear, supports multi-step ahead prediction, and works
very well under time variant and non-stationary scenarios such
as single server queuing systems under time varying mean arrival
rate. Performance of the proposed predictor is evaluated through
simulation. Simulations show that the proposed predictor is able
to predict the response times of single server queuing systems in
multi-step ahead with very good precision represented by very
small mean absolute and mean squared prediction errors. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/2018675
- author
- Amani, Payam LU ; Kihl, Maria LU and Robertsson, Anders LU
- organization
- publishing date
- 2011
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- [Host publication title missing]
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- The 16th IEEE Symposium on Computers and Communications (ISCC)
- conference location
- Kerkyra (Corfu), Greece
- conference dates
- 2011-06-28 - 2011-07-01
- external identifiers
-
- wos:000298614900165
- scopus:80052720751
- project
- Broadband Communications: Modelling and Control of Server Systems
- LCCC
- language
- English
- LU publication?
- yes
- id
- 0fe3f77a-16f0-46d6-846e-5d484e181f0c (old id 2018675)
- date added to LUP
- 2016-04-04 12:13:46
- date last changed
- 2022-01-29 23:08:47
@inproceedings{0fe3f77a-16f0-46d6-846e-5d484e181f0c,
abstract = {{Multi-step ahead response time prediction of CPU<br/><br>
constrained computing systems is vital for admission control,<br/><br>
overload protection and optimization of resource allocation in<br/><br>
these systems. CPU constrained computing systems such as web<br/><br>
servers can be modeled as single server queuing systems. These<br/><br>
systems are stochastic and nonlinear. Thus, a well-designed nonlinear<br/><br>
prediction scheme would be able to represent the dynamics<br/><br>
of such a system much better than a linear scheme. A nonlinear<br/><br>
autoregressive neural network with exogenous inputs based<br/><br>
multi-step ahead response time predictor has been developed.<br/><br>
The proposed estimator has many promising characteristics that<br/><br>
make it a viable candidate for being implemented in admission<br/><br>
control products for computing systems. It has a simple structure,<br/><br>
is nonlinear, supports multi-step ahead prediction, and works<br/><br>
very well under time variant and non-stationary scenarios such<br/><br>
as single server queuing systems under time varying mean arrival<br/><br>
rate. Performance of the proposed predictor is evaluated through<br/><br>
simulation. Simulations show that the proposed predictor is able<br/><br>
to predict the response times of single server queuing systems in<br/><br>
multi-step ahead with very good precision represented by very<br/><br>
small mean absolute and mean squared prediction errors.}},
author = {{Amani, Payam and Kihl, Maria and Robertsson, Anders}},
booktitle = {{[Host publication title missing]}},
language = {{eng}},
publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
title = {{Multi-step ahead response time prediction for single server queuing systems}},
url = {{https://lup.lub.lu.se/search/files/5957669/2018679.pdf}},
year = {{2011}},
}