Abstract
This paper describes an application-specific processor (add-on chips) that helps computers work harder without consuming massive amounts of energy. If the problem is the power demand of general-purpose processors, the answer may lie in the application-specific acceleration, a technique being tried out in supercomputers. UK-based Clearspeed Technology reworked a chip design developed for PC graphics into a massively parallel floating-point calculation workhouse. Another main option for acceleration is to use a field-programmable gate array (FPGA), whose logic circuits are not hardwired but defined by loading a configuration memory.
Full Text
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call