Architecture, algorithms and applications for future generation supercomputers

We outline a hierarchical architecture for machines capable of over 100 teraOps in a 10 year time-frame. The motivating factors for the design are technological feasibility and economic viability. The envisioned architecture can be built largely from commodity components. The development costs of the machine will therefore be shared by the market. To obtain sustained performance from the machine, we propose a heterogeneous programming environment for the machine. The programming environment optimally uses the power of the hierarchy. Programming models for the stronger machine models existing at the lower levels are tuned for ease of programming. Higher levels of the hierarchy place progressively greater emphasis on locality of data reference. The envisioned machine architecture requires new algorithm design methodologies. We propose to develop hierarchical parallel algorithms and scalability metrics for evaluating such algorithms. We identify three important application areas: large scale numerical simulations, problems in particle dynamics and boundary element methods, and emerging large-scale applications such as data-mining. We briefly outline the process of hierarchical algorithm design for each of these application areas.