SWEB: Towards a Scalable World Wide Web Server on Multicomputers

We investigate the issues involved in developing a scalable World Wide Web(WWW) server on a cluster of workstations and parallel machines, using theHypertext Transport Protocol (HTTP). The main objective is to strengthen theprocessing capabilities of such a server by utilizing the power ofmulticomputers to match huge demands in simultaneous access requests from theInternet. We have implemented a system called SWEB on a distributed memorymachine, the Meiko CS-2, and networked SUN and DEC workstations. Thescheduling component of the system actively monitors the usages of CPU, I/Ochannels and the interconnection network to effectively distribute HTTPrequests across processing units to exploit task and I/O parallelism. Wepresent the experimental results on the performance of this system. Ourresults indicate that the system delivers good performance on multi-computersand obtains significant improvements over other approaches.