Performance of Inverted Indices in Distributed Text Document Retrieval Systems
Anthony Tomasic, Héctor García-Molina · 1993
The performance of distributed text document retrieval systems is strongly influenced by the organization of the inverted index. This paper compares the performance impact on query processing of various physical organizations for inverted lists. We present a new probabilistic model of the database and queries. Simulation experiments determine which variables most strongly influence response time and throughput. This leads to a set of design trade-offs over a wide range of hardware configurations and new parallel query processing strategies. 1 Introduction Full text databases of newspaper articles, journals, legal documents etc. are readily available. These databases are rapidly increasing in size as the cost of digital storage drops, as more source documents are available in electronic form, and as optical character recognition becomes commonplace. At the same time, there is a rapid increase in the number of users and queries submitted to such text retrieval systems. One reason is tha...