Forest logging: A trace-based analysis of large rule-based computations

Knowledge representation systems based on thewell-founded semantics can offer the degree of scalabilityrequired for semantic web applications and make use of ex-pressive semantic features such as Hilog, frame-based rea-soning, and defeasibility theories. Such features can be com-piled into Prolog tabling engines that have good support forindexing and memory management. However, due both tothe power of the semantic features and to the declarativestyle typical of knowledge representation rules, the resourcesneeded for query evaluation can be unpredictable. In sucha situation, users need to understand the overall structureof a computation and examine problematic portions of it.This problem, of profiling a computation, differs from debug-ging and justification which address why a given answer wasor wasn’t derived, and so profiling requires different tech-niques. In this paper we present a trace-based analysis tech-nique called forest logging which has been used to profilelarge, heavily tabled computations. In forest logging, criti-cal aspects of a tabled computation are logged; afterwardsthe log is loaded and analyzed. As implemented in XSB, for-est logging slows down execution of practical programs by aconstant factor that is often small; and logs containing tensor hundreds of millions of facts can be loaded and analyzedin minutes.