Performance Guarantee in a New Hybrid Push-Pull Scheduling Algorithm

The rapid growth of web services has already given birth to a set of data dissemination applications. Efficient scheduling techniques are necessary to endow such applications with advanced data processing capability. In this paper we have effectively combined broadcasting of very popular (push) data and dissemination of less popular (pull) data to develop a new hybrid scheduling scheme. The separation between the push and the pull data is called cut-off point. The clients send their request to the server, which ignores the request for the push items but queues the requests for the pull items. At every instance of time, the item to be broadcast is designated by applying a pure-push scheduling. On the other hand, the item to be pulled is the one stored in the pull-queue having the highest number of pending requests. Fixed a value for the cut-off point, an analytic model, validated by simulation, evaluates the average system performance. On the top of that, the major novelty of our system lies in its capability of offering a performance guarantee to the clients. For each request, the client specifies, along with the data item requested, the maximum interval of time it can afford to wait before its request will be served. Based on this particular access-time and on the analytic model, our hybrid system computes the cut-off point to fulfill the specific need of the client. The system offers a great flexibility for clients. It improves significantly upon a pure push system and some existing hybrid systems in terms of average waiting time spent by a client.