Allocating data objects to multiple sites for fast browsing of hypermedia documents

Many World Wide Web applications require access to and the transfer and synchronization of large multimedia data objects (MDOs), such as audio, video and images, across the communication network. The transfer of large MDOs contributes to the response time observed by the end users. As the end users expect strict adherence to response time constraints, the problem of allocating these MDOs so as to minimize response time becomes very challenging. The problem becomes more complex in the context of hypermedia documents (Web pages), wherein these MDOs need to be synchronized during presentation to the end users. Since the basic problem of data allocation in distributed database systems is NP-complete, a need exists to pursue and evaluate solutions based on heuristics for generating near-optimal MDO allocations. In this paper, we (i) conceptualize this problem by using a navigational model to represent hypermedia documents and their access behavior by end users, (ii) formulate the problem by developing a base case cost model for response time, (iii) design two algorithms to find near-optimal solutions for allocating MDOs of the hypermedia documents while adhering to the synchronization requirements, and (iv) evaluate the trade-off between the time complexity to obtain the solution and the quality of the solution by comparing the algorithms' solutions with an exhaustive solution over a set of experiments.