Response time in large, inverted file document retrieval systems is determined by time required to access files of document identifiers on disk and process a Boolean search request. A specialized computer system has been devised that can perform a complicated sample search involving 70 terms and over 60,000 document references 12 to 60 times faster than a conventional machine. Many small searches can be processed concurrently with little effect on system performance. The system can be realized with currently available technology and has been tried in numerous simulations involving various system configurations and other factors. (SK)