2016 Benjamin Franklin Medal in Computer and Cognitive Science presented to Yale N. Patt, Ph.D.

摘要

Modern computing is based in large part on an important abstraction called the von Neumann model. In this model, the program consists of a series of instructions stored in memory. The computing hardware performs the program by executing the instructions in sequential order. This basic model constitutes a contract that allows software developers to construct programs based on a simple, easy to understand execution model while delegating the details of the execution of those instructions to the hardware.Yale Patt and his colleagues have been responsible for pioneering new approaches to the design of computer systems that achieve increased performance by automatically identifying instructions that can be performed in parallel. This approach allows us to preserve the model of sequential execution while reaping the performance benefits associated with concurrency.