REPRODUCE 2014 - Workshop on Reproducible Research Methodologies

Computer science and engineering research fields increasingly rely on numerous ad hoc methods to explore research breakthroughs, particularly during empirical and statistical analysis, modeling, optimization and simulation of complex computer systems. These ad hoc methods are utilized due to a variety of factors including problem complexity and size, speed of advancement and return on investment, cost of designing prototypes, and minimal access to state-of-the-art fabrication. However, lack of a common experimental methodology, and lack of simple and unified mechanisms, tools and repositories to preserve and exchange the whole experimental setup including all past research artifacts makes it excessively challenging or even impossible to accurately reproduce experimental results for evaluation and future advancement.
This workshop is an interdisciplinary forum for academic and industrial researchers, practitioners and developers in computer engineering to discuss ideas, experience, trustable and reproducible research methodologies, practical techniques, tools and repositories. Detailed topics of interest include but are not limited to:
Techniques for experimental reproducibility
Fidelity validation of experiments and tools
Reproducible benchmarking methodologies
Scientific method for computing experiments
Techniques to effectively leverage emulation
Scalable HPC computing in experimentation
Metrics quantifying performance vs. fidelity
Reproducible and extensible methodologies
Data-mining, crowdsourcing analysis methods
Interdisciplinary experimental methods
Tool, benchmark, experiment repositories
Program and architecture auto-tuning
Reproducible simulation environments
Experimental documentation methods
Techniques to preserve/share experiments
Methods to effectively analyze experiments
Extracting metadata from experiments
Methods for protecting intellectual property