A data analyst is executing a shell script across all notebooks linked to a cluster, but the script's lengthy command set is causing significant delays. As a data engineer, what advice would you offer to expedite the process?