| It is a real pleasure and honor for us to present you this book titled Performance Evaluation and Benchmarking. Performance evaluation and benchmarking is at the heart of computer architecture research and development. Without a deep understanding of benchmarks’ behavior on a microprocessor and without efficient and accurate performance evaluation techniques, it is impossible to design next-generation microprocessors. Because this research field is growing and has gained interest and importance over the last few years, we thought it would be appropriate to collect a number of these important recent advances in the field into a research book. This book deals with a large variety of state-of-the-art performance evaluation and benchmarking techniques. The subjects in this book range from simulation models to real hardware performance evaluation, from analytical modeling to fast simulation techniques and detailed simulation models, from single-number performance measurements to the use of statistics for dealing with large data sets, from existing benchmark suites to the conception of representative benchmark suites, from program analysis and workload characterization to its impact on performance evaluation, and other interesting topics. We expect it to be useful to graduate students in computer architecture and to computer architects and designers in the industry.
This book was not entirely written by us. We invited several leading experts in the field to write a chapter on their recent research efforts in the field of performance evaluation and benchmarking. We would like to thank Prof. David J. Lilja from the University of Minnesota, Prof. Tom Conte from North Carolina State University, Prof. Brad Calder from the University of California San Diego, Prof. Chita Das from Penn State, Prof. Brinkley Sprunt from Bucknell University, Alex Mericas from IBM, and Dr. Kishore Menezes from Intel Corporation for accepting our invitation. We thank them and their co-authors for contributing. Special thanks to Dr. Joshua J. Yi from Freescale Semiconductor Inc., Paul D. Bryan from North Carolina State University, Erez Perelman from the University of California San Diego, Prof. Timothy Sherwood from the University of California at Santa Barbara, Prof. Greg Hamerly from Baylor University, Prof. Eun Jung Kim from Texas A&M University, Prof. Ki HwanYum from the University of Texas at San Antonio, Dr. Rumi Zahir from Intel Corporation, and Dr. Susith Fernando from Intel Corporation for contributing. Many authors went beyond their call to adjust their chapters according to the other chapters. Without their hard work, it would have been impossible to create this book. We hope you will enjoy reading this book. |