Advanced

Some topics in web performance analysis

Cao, Jianhua LU (2004)
Abstract
This thesis consists of four papers on web performance analysis. In the first paper we

investigate the performance of overload control through queue length for two different web

server architectures. The simulation result suggests that the benefit of request prioritization

is noticeable only when the capacities of the sub-systems match each other. In the second

paper we present an M/G/1/K*PS queueing model of a web server. We obtain closed form

expressions for web server performance metrics such as average response time, throughput

and blocking probability. The model is validated through real measurements. The third

paper studies a queueing system with a load balancer and a... (More)
This thesis consists of four papers on web performance analysis. In the first paper we

investigate the performance of overload control through queue length for two different web

server architectures. The simulation result suggests that the benefit of request prioritization

is noticeable only when the capacities of the sub-systems match each other. In the second

paper we present an M/G/1/K*PS queueing model of a web server. We obtain closed form

expressions for web server performance metrics such as average response time, throughput

and blocking probability. The model is validated through real measurements. The third

paper studies a queueing system with a load balancer and a pool of identical FCFS queues

in parallel. By taking the number of servers to infinite, we show that the average waiting

time for the system is not always minimized by routing each customer to the expected

shortest queue when the information used for decision is stale. In the last paper we

consider the problem of admission control to an M/M/1 queue under periodic observations

with average cost criterion. The problem is formulated as a discrete time Markov decision

process whose states are fully observable. A proof of the existence of the average optimal

policy by the vanishing discounted approach is provided. We also show that the optimal

policy is nonincreasing with respect to the observed number of customers in the system. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Thesis
publication status
published
subject
pages
84 pages
publisher
Lund Institute of Technology
language
English
LU publication?
yes
id
4c75f6d2-b8ff-415d-9366-dcaad3262701 (old id 532399)
date added to LUP
2007-10-03 14:46:17
date last changed
2016-09-19 08:44:59
@misc{4c75f6d2-b8ff-415d-9366-dcaad3262701,
  abstract     = {This thesis consists of four papers on web performance analysis. In the first paper we<br/><br>
investigate the performance of overload control through queue length for two different web<br/><br>
server architectures. The simulation result suggests that the benefit of request prioritization<br/><br>
is noticeable only when the capacities of the sub-systems match each other. In the second<br/><br>
paper we present an M/G/1/K*PS queueing model of a web server. We obtain closed form<br/><br>
expressions for web server performance metrics such as average response time, throughput<br/><br>
and blocking probability. The model is validated through real measurements. The third<br/><br>
paper studies a queueing system with a load balancer and a pool of identical FCFS queues<br/><br>
in parallel. By taking the number of servers to infinite, we show that the average waiting<br/><br>
time for the system is not always minimized by routing each customer to the expected<br/><br>
shortest queue when the information used for decision is stale. In the last paper we<br/><br>
consider the problem of admission control to an M/M/1 queue under periodic observations<br/><br>
with average cost criterion. The problem is formulated as a discrete time Markov decision<br/><br>
process whose states are fully observable. A proof of the existence of the average optimal<br/><br>
policy by the vanishing discounted approach is provided. We also show that the optimal<br/><br>
policy is nonincreasing with respect to the observed number of customers in the system.},
  author       = {Cao, Jianhua},
  language     = {eng},
  note         = {Licentiate Thesis},
  pages        = {84},
  publisher    = {Lund Institute of Technology},
  title        = {Some topics in web performance analysis},
  year         = {2004},
}