Collaborative Research: SMA: Accurate Temperature Measurement Infrastructure and Methodology for Power, Variability, and Reliability Analysis

University Of Rochester, Rochester NY

Investigators

Abstract

Simulation environments are an indispensable tool in the design, prototyping, performance evaluation, and analysis of computer systems. Simulator must be able to faithfully reflect the behavior of the system being analyzed. To ensure the accuracy of the simulator, it must be verified and determined to closely match empirical data. Modern processors provide enough performance counters to validate the majority of the performance models; nevertheless, the information provided is not enough to validate power consumption, thermal models, and process variability. Temperature and power consumption are first order design parameters for modern high performance architectures. High operational temperatures and large power consumption present possible limits to performance and manufacturability. While temperature and power are dominant factors on processor design, process variability is becoming another major constraint. In order to address some of the difficulties associated with the validation of power, thermal models, and process variability, we propose to use an infrared measurement setup to capture run-time power consumption and thermal characteristics of modern chips. A detailed thermal model can use the measured temperatures to generate a power detailed power consumption. The same thermal measurements will be used to measure process variability from of-the-shelf processors. In addition, the resulting infrastructure will be use to develop a verified thermal simulation infrastructure and process variability models.