A Productive Time Length-Based Method for Multi-tenancy-Oriented Service Usage Metering and Billing

Abstract

Currently, most of the service computing-related platforms provide services free of charge or simply charge users fees based on storage occupation or processor capacity as tradition. For the services themselves, which almost the only functional elements in a service computing environment [1,2], the usage metering and billing is still a not well-solved but important problem. To solve the Multi-tenancy–oriented [3] service usage metering and billing problem, this paper presents a productive time length-based method, compared to the traditional method for multi-tenancy -oriented service usage metering and billing, the productive time length-based method is more discriminative and can improve the accuracy of the metering process. It can avoid the mis-metering of services that return failed or contribute little to the final output user-obtained.