A PetaFLOPS computing system will be feasible in two decades and will be important, perhaps even critical, to key applications at that time. Its practicality will rely, in part, on continued advance of the semiconductor industry both in speed enhancement and in cost reduction through improved fabrication processes. While no paradigm shift is required in systems architecture, active latency management will be essential, requiring a very high degree of fine-grain parallelism and the mechanisms to exploit it. One tenable path in device technologies is a mix of technologies including semiconductor for main memory, optics for interprocessor (and perhaps interchip) communications and secondary storage, and possibly superconducting for very high clock rate and very low power processor logic. Effectiveness and applicability will rest on dramatic per device cost reduction and innovative approaches to system software and programming methodologies. Near term studies are required to refine these findings through more detailed examination of system requirements and technology extrapolation.