Tools for Mapping, Load Balancing and Monitoring in the LOGFLOW Parallel Prolog Project

LOGFLOW is an all-solution parallel logic programming system able to exploit OR-parallelism and pipeline AND-parallelism of Prolog programs. The LOGFLOW project is intended to implement Prolog in massively parallel distributed memory multicomputers. Porting LOGFLOW to a workstation cluster resulted in a variant of LOGFLOW called WS-LOGFLOW. Implementation of LOGFLOW both on multi-transputers and on workstation clusters required to solve many parallel software engineering problems like load balancing, mapping, monitoring, debugging, etc. In this paper the load balancing, mapping and monitoring techniques applied in multi-transputer based LOGFLOW systems have been described. The concept of WS-LOGFLOW and its Distributed Token Scheduler scheme for load balancing is also explained in the paper.