A Python file is ready to go into production and the client wants to use the cheapest but most efficient type of cluster possible. The workload is quite small, only processing 10GBs of data with only simple joins and no complex aggregations or wide transformations.

Which cluster meets the requirement?