DNS-based load balancing with Cache for geographically distributed Web server systems
Fouzi Semchedine, Louiza Bouallouche‐Medjkoune, Omar Sayeh, Sofiane Ayoub, Djamil Aı̈ssani · 2014
The scalability and the availability of the Web can be provided by distributing the Web servers (Web Cluster) where the client requests must be balanced among these Web servers in order to improve the performance. The DNS-based load balancing mechanism was becoming one of the most challenges for these last years. This paper investigates a proposition of a new variant of the DNS-based load balancing technique DLB-TS (Dynamic Load Balancing Based on Task Sizes). We propose, in this version, a new technique called DLB-TSC (Dynamic Load Balancing Based on Task Sizes with Cache memory) in which we improve the DLB-TS technique by introducing a cache mechanism in order to cache the most used Web files and reduce the variability in request sizes. Simulation results show that DLB-TSC improves the performance more than the standard DLB-TS technique in a wide range.