This paper presents a group of multiple-way graph (with weighted nodes
and edges) partitioning algorithms based on a 2-stage constructive-an
d-refinement mechanism. The graph partitions can be used to control al
location of program units to distributed processors in a way that mini
mizes the completion time and for design automation applications. In t
he constructive stage, 4 clustering algorithms are used to construct r
aw partitions, the second refinement step first adjusts the cluster nu
mber to the processor number and then iteratively improves the partiti
oning cost by employing a Kernighan-Lin based heuristic. This approach
represents several extensions to the state-of-the-art methods. A perf
ormance comparison of the proposed algorithms is given, based on exper
iment results.