Output details
11 - Computer Science and Informatics
University of Edinburgh
An Empirical Architecture-Centric Approach to Microarchitectural Design Space Exploration
<02>Originality: This work presents a novel machine-learning model that can be used to speed up the design process of new microprocessors.
Significance: When developing new microprocessors, designers have to run thousands of benchmarks in order to assess the performance characteristics of their design. This process might take days or even weeks to complete as thousands of detailed simulations need to be run. The techniques presented in this work can reduce by an order of magnitude the time required to run these simulations.
Rigour: Based on experimental results from the state-of-the-art (at the time) simulator used both in academia and industry.