1.0.0.0.5 Fast Loading of Code

A further goal of XSB is to provide in implementation engine for both logic programming and for data-oriented applications such as in-memory deductive database queries and data mining [48]. One prerequisite for this functionality is the ability to load a large amount of data very quickly. We have taken care to code in C a compiler for asserted clauses. The result is that the speed of asserting and retracting code is faster in XSB than in any other Prolog system of which we are aware, even when some of the sophisticated indexing mechanisms described above are employed. At the same time, because asserted code is compiled into SLG-WAM code, the speed of executing asserted code in XSB is faster than that of many other Prologs as well. We note however, that XSB does not follow the ISO-semantics of assert [40].