A second and revised version of the Parallel Prolog Abstract Machine (PPAM) is described in the paper. PPAM-2 is suitable for implementing Prolog on communicating process architecture-type, highly parallel computers with regular interconnection network. The parallel computational model behind PPAM-2 is an extended cellular-dataflow model, whose main features have been published elsewhere. Based on the model, Prolog programs are transformed into a dataflow search graph representing the static structure of the program. The paper explains in detail how the Prolog programs can be transformed into the dataflow graph form. The dynamic behavior of the program is determined by the extended transition functions of the operators. By modifying the extended transition functions different kinds of parallelism can be achieved during the execution of the PPAM-2 code. In the paper an informal description is given of the transition functions realising OR-parallelism and pipelined AND-parallelism. A multi-transputer implementation of PPAM-2 has been investigated in the paper. First the chosen interconnection topology of the multi-transputer-based machine is described, then the method for mapping the dataflow search graph of Prolog programs on to the chosen topology. Finally, the realisation techniques of the dataflow operations on transputers are overviewed.
ASJC Scopus subject areas
- Computer Science(all)