Dual Scaling VMs and Queries: Cost-Effective Latency Curtailment | IEEE Conference Publication | IEEE Xplore