Admission Control of the Apache Web Server.pdf

Admission Control of the Apache Web Server M. Andersson, M. Kihl, A. Robertsson and B. Wittenmark. Abstract Web sites are exposed to high rates of incoming requests. The servers may become overloaded during temporary traffic peaks when more requests arrive than the server is designed for. An admission control mechanism rejects some requests whenever the arriving traffic is too high and thereby maintains an acceptable load in the system. This paper presents how admission control mechanisms can be designed with a combination of queueing theory and control theory. In this paper we model an Apache web server as a G/G/1-system and then design a PI-controller, commonly used in automatic control, for the server. The controller has been implemented as a module inside the Apache source code. Measurements from the laboratory setup show how robust the implemented controller is, and how it correspond to the results from the theoretical analysis. I. INTRODUCTION One problem with web servers is that they are sensitive to overload. The servers may become overloaded during temporary traffic peaks when more requests arrive than the server is designed for. Because overload usually occurs rather seldom, it is not economical to overprovision the servers for these traffic peaks, instead admission control mechanisms can be implemented in the servers. The admission control mechanism rejects some requests whenever the arriving traffic is too high and thereby maintains an acceptable load in the system. Traditionally, server utilization or queue lengths have been the variables mostly used in admission control schemes. For web servers, the main objective of the control scheme is to protect it from overload. As long as the average server utilization or queue length is below a certain level, the response times are low. One well-known controller in automatic control is the PID-controller, which enables a stable control for many types of systems (see, for example ?str?m, [1]). In order to get the