The dataPath directory is meant to be a shared data repository for all users to access data on the cluster. It can be a character vector of paths, and is intended to be places where common data can be stored on each node and thus available to many users. The workingDir is used to specify a personal R working directory and will default to the standard Windows user directory.