A parallel execution model for Chronolog

Temporal logic has been widely used as a formalism for program specification and verification [9, 12, 13], modelling temporal databases [1, 3, 14] and reasoning about time [8, 19, 20]. In temporal logic, the meanings of formulae depend on an implicit time parameter and elements from different moments in time can be combined through the use of temporal operators. Therefore, temporal logic can model timedependent and dynamic properties of certain problems in the real-world in a natural way. Recently, several researchers have suggested that temporal logiccan be used as a programming language; see [15] for a comprehensive survey. However, there have been a few attempts at implementing temporal logic languages. Some of the early suggested implementations, which are based on translation into standard logic programs [21] or by meta-interpretation [7], cannot exploit parallelism inherent in logic programs and context-parallelism which is offered by temporal logic programs. Chronolog [16, 17] is a temporal extension of logic programming, based on a linear-time temporal logic [4] with an unbounded future in which the set of natural numbers models the collection of moments in time. For the temporal logic programming language Chronolog, we have outlined a parallel execution model, called CHEM (CHronolog Execution Model) [10]. The model is based on dataflow computation. Chronolog(Z}) is an extension of Chronolog based on a linear-timetemporal logic with unbounded past and future.By adding ‘choice predicates’ to Chronolog(Z), we can obtain exactly one answer to a given goal when we want to model dataflow style of stream-oriented computations. In order to exploit inherent parallelismin Chronolog(Z), this paper,