Adaptive and scalable allocation of data-objects in the web

Abstract

In this paper we address the problem of designing allocation schemas of large database-objects in a Web environment that may be exposed to significant changes in usage and access patterns and scaling of data. In these circumstances, if the design is not adjusted to the new changes, the system can undergo a severe degradation in data access costs and response time. In order to solve this problem, we propose a formal model to generate a new database-object allocation. The model uses current state information of the system and usage statistical data collected during a given period, and adapts the system to the new usage patterns so as to minimize communication costs. Implicitly the mathematical model handles database-object migration and scaling of the number of sites. Model tests have been conducted with satisfactory and promising results. The principles used in the model can be applied to other models for the optimization of Web resources. © Springer-Verlag Berlin Heidelberg 2003.