Enhanced ATAMM for increased throughput performance of multicomputer data flow architectures
Robert L. Jones, Paul J. Hayes, ASA ANDREWS, S. Som, John W. Stoughton, Roland R. Mielke · 2002
The algorithm to architecture mapping model (ATAMM) is a Petri-net-based model which provides a strategy for periodic execution of a class of real-time algorithms on multicomputer data flow architectures. The problem domain of particular interest is the execution of large-grained, decision-free algorithms on homogeneous processing elements. Design techniques are discussed and performance measurements are defined. A multiple-graph execution strategy is shown to increase throughput performance. It is shown that the same increase in performance is attainable with minor modifications to the existing ATAMM.>