Fooling the Masses with Performance Results: Old Classics & Some New Ideas
ABSTRACT
This talk will be done using the slides for a presentation that was given by Dr. Wellein @ ISC 2013.
The presentation is based on the classic article "Twelve Ways to Fool the Masses When Giving Performance Results on Parallel Computers" by David H. Bailey and it describes the ways in which research papers in parallel and high performance computing present data to artificially boost performance rates and to present these results in the "best possible light."
Index Terms—computational biology; finite automaton; graph algorithms; hardware acceleration; motif detection.
BIO
Jee Choi is an E.C.E. PhD student with Prof. Richard Vuduc. My main field of research is High Performance Computing and my past research topics include optimizing and modeling Sparse Matrix Vector (SpMV), the Fast Multipole Method (FMM) and Lattice Quantum Chromodynamics (LQCD) on accelerators.
My most recent research topic is modeling power and energy for high performance systems via the "energy roofline model."