We develop systems that retroactively facilitate computational reproducibility
RaaS is a web server-based application that provides retroactive computational reproducibility. This tool analyzes computational experiments and generates a Docker image with the scripts, data, and computational environment necessary to reproduce the original analysis. During the build process of the image, we collect provenance while executing the analyses. The provenance serves as a record of what happened and can provide useful information for understanding the results. Future reproducers of the experiment can run the scripts or analyze the results and provenance generated during the build process without re-executing them. Additionally, they can tweak the experiment or write their own analysis for the same data to see how the results may vary. Since Docker images are immutable, reproducers can experiment as much as they want in the container without fear of losing the original results.
Currently, we are finalizing the implementation of a backend centered around isolating language-independent code. Future users will be able to implement support for new programming languages quickly following a standard template. Below is the status of the various languages RaaS currently supports:
Systopia lab is supported by a number of government and industrial sources, including Cisco Systems, the Communications Security Establishment Canada, Intel Research, the National Sciences and Engineering Research Council of Canada (NSERC), Network Appliance, Office of the Privacy Commissioner of Canada, and the National Science Foundation (NSF).