Designing Large Hierarchical Multiprocessor Systems under Processor, Interconnection, and Packaging Advancements

A general framework for architectural design of large hierarchical multiprocessor systems under rapidly changing packaging, processor, and interconnection technologies is presented. In recent years processor boards with larger area (A) and greater pinouts are becoming feasible. Board interconnection technology has advanced from peripheral connections O(\sqrt A ) to elastomeric surface connections 0(A). As processor and interconnection technology grows, there is a varying demand on the interconnection network of the system. The proposed framework is capable of taking into account all these changes in technologies and, depending on a given set of technological parameters, derive the most optimum topology. The framework is illustrated by considering the design problem of the currently popular class of k-ary n-cube cluster-c scalable architectures.