A storage administrator has mounted an NFS v3 datastore on multiple ESXi hosts connected to a Cisco Nexus 9000 leaf pair. During month-end batch jobs, a low-priority reporting VM consumes most of the NFS array's throughput, causing a business-critical database VM on the same datastore to experience high read latency. The admin wants to ensure the database VM receives proportional access to storage resources during contention without changing the physical storage or network fabric. Which mechanism should be used?