In recent years there has been a growing effort to capture and record script or simulation executions in scientific computing. In the litterature, there is a convergence towards what the record should contain to better address reproducibility, management and reliability. Tools such as Sumatra, Reprozip and CDE address different aspects of the problem from simple record keeping to generating a complete bundle of the environment at each execution. A complete simulation management tool (SMT) must address a number of key functions including record keeping, live job inspection, meta-data capture, environment capture, as well as possibly bundling the environment for later use. In addition, a cloud platform to augment SMTs that displays and analyzes metadata for each simulation record has the potential to greatly improve the practice of simulation management in scientific computing.
CoRR (Cloud of Reproducible Records) is a cloud platform aimed at augmenting existing SMTs such as Sumatra. CoRR provides services for both storing and viewing simulation records using any SMT that integrates its API. The intent of CoRR is to be as robust and record scheme agnostic as possible so that changes to the SMTs, backend model, API and frontend do not break the interactions between each element or backwards compatibility with existing record sets. The Python web framework, Flask is used to generate the backend endpoints while The frontend is entirely HTML5, JavaScript and CSS based and generates the data object model based solely on the record available.