In a distributed computing environment, you are working with a large dataset that needs to be processed in parallel. Which data structure would be most appropriate for representing the relationships between the data elements in this scenario?