Multimedia Streams Retrieval in Distributed Systems Using Learning Automata

Current academic Internet environment has enabled fast transfers of huge amounts of data, and has made high quality multimedia and collaborative applications a reality. This article describes a model for distributed multimedia retrieval which performs the retrieval of different multimedia to a variety of clients using learning automata algorithm named LAGridMSS. LAGridMSS allocates a proportion of bandwidth of each node for sending a specified file, and then applies learning automata to allocate packets of files to each node that contains the context. The files’ popularity and the remainder bandwidth of nodes are two main factors here for allocating packets to each node. Simulation results show improvements in proposed model, LAGridMSS, in some QoS factors such as delay, jitter, and reliability compared to previous multimedia retrieval system such as GridMSS and GridMedia.