A comparison of concatenated and superimposed code word surrogate files for very large data/knowledge bases

Abstract

Surrogate files are very useful as an index for very large knowledge bases to support multiple logic programming inference mechanisms because of their small size and simple maintenance requirement. In this paper, we analyse the superimposed code word (SCW) and concatenated code word (CCW) surrogate file techniques in terms of storage space and time to answer queries in various cases. One of the most important results of our analysis is that the size and the query response time of the CCW is smaller than those of the SCW when the average number of arguments specified in a query is small. It is also shown that most of the query response time is used for the surrogate file processing when the extensional database is very large. Therefore, if we use a special architecture to speed up the surrogate file processing, the total query response time can be reduced considerably.