Modeling Web Server Traffic with Session-Based Arrival Streams

Authors:
Laurence Hoflack;Stijn Vuyst;Sabine Wittevrongel;Herwig Bruneel
Affiliations:
TELIN Department, SMACS Research Group, Ghent University, Gent, Belgium B-9000;TELIN Department, SMACS Research Group, Ghent University, Gent, Belgium B-9000;TELIN Department, SMACS Research Group, Ghent University, Gent, Belgium B-9000;TELIN Department, SMACS Research Group, Ghent University, Gent, Belgium B-9000
Venue:
ASMTA '08 Proceedings of the 15th international conference on Analytical and Stochastic Modeling Techniques and Applications
Year:
2008

Citing 8
Cited 1

Packet delay and queue length for statistical multiplexers with low-speed access lines

Computer Networks and ISDN Systems
Internet Web servers: workload characterization and performance implications

IEEE/ACM Transactions on Networking (TON)
Self-similarity in World Wide Web traffic: evidence and possible causes

IEEE/ACM Transactions on Networking (TON)
Correlation effects in ATM queues due to data format conversions

Performance Evaluation
Discrete-Time Models for Communication Systems Including ATM

Discrete-Time Models for Communication Systems Including ATM
Mean value and tail distribution of the message delay in statistical multiplexers with correlated train arrivals

Performance Evaluation
Performance analysis of a discrete-time queuing system with a correlated train arrival process

Performance Evaluation
System Content and Packet Delay in Discrete-Time Queues with Session-Based Arrivals

ITNG '08 Proceedings of the Fifth International Conference on Information Technology: New Generations

Analysis of Discrete-Time Buffers with General Session-Based Arrivals

ASMTA '09 Proceedings of the 16th International Conference on Analytical and Stochastic Modeling Techniques and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we consider a discrete-time queueing model with infinite waiting room and one single output channel. Users from an infinite user population can start and end sessions during which they are active and send packets to a buffer system. Each active user generates a random but strictly positive number of packets per time slot. The queueing model is applied to study the traffic of a web server, where each session corresponds to one file download. This approach has the advantage of leading to analytical expressions for the main performance measures (buffer occupancy, delay, ...) of the output buffer of the web server. Specifically, in this paper, a closed-form expression for the mean session delay is obtained. We then apply the model to a web server, based on a trace of actual web traffic. Using the model, we can rapidly study the influence of the different system parameters and compare various alternatives.