A Day Late and a Dollar Short: The Case for Research on Cloud Billing Systems

Cloud computing platforms such as Amazon Web Services, Google Compute Engine, and Rackspace Public Cloud have been the subject of numerous measurement studies considering performance, reliability, and cost efficiency. However, little attention has been paid to billing. Cloud providers rely upon complex, large-scale billing systems that track customer resource usage at fine granularity and generate bills reflecting measured usage. However, it is not known how visible such usage is to customers, and how closely provider charges correspond to customers’ view of their resource usage.

We initiate a study of cloud billing systems, focusing on Amazon EC2, Google Compute Engine, and Rackspace, and uncover a variety of issues, including: inherent difficulties in predicting charges; bugs that lead to free CPU time on EC2 and over-charging for storage in Rackspace; and long and unpredictable billing-update latency. Our measurements motivate further study on billing systems, and so we conclude with a brief discussion of open questions for future work.