The Test and Evaluation Uses of Heterogeneous Computing: GPGPUs and Other Approaches

摘要

The test and evaluation community faces conflicting pressures: Provide more computing power and reduce electrical power requirements, both on the range and in the laboratory. The authors present some quantifiable benefits from the implementation of General Purpose Graphics Processing Units (GPGPUs) as heterogeneous processors. This produces power, space, cooling, and maintenance benefits that they have documented. Other efforts in the field of power reduction techniques will be outlined, e.g., the efficient low-power microprocessor approach of Prof. William Dally and IBM's well-publicized Blue Gene project. The utility of all of these techniques for the test and evaluation community is assessed. The authors will report on several aspects of their experience with GPGPUs: programmability, performance of codes implemented in several areas of computational science, and the compute power per unit of electrical consumption. An overview of code design and implementation approaches is discussed.