Developing a Cloud Computing Charging Model for High-Performance Computing Resources

Abstract

This paper examines the economics of cloud computing charging from the perspective of a supercomputing resource provider offering its own resources. To evaluate the competitiveness of our computing center with cloud computing resources, we develop a comprehensive system utilization charging model similar to that used by Amazon EC2 and apply the model to our current resources and planned procurements. For our current resource, we find that charging for computational time may be appropriate, but that charging for data traffic between the supercomputer and the storage/front-end systems would result in negligible additional revenue. Similarly, charging for data storage capacity at currently typical commercial rates yields insufficient revenue to offset the acquisition and operation of the storage. However, when we extend the analysis to a capacity cluster scheduled for deployment in the first half of 2010 that will be made available to users through batch, Grid, and cloud interfaces, we find that the resource will be competitive with current and anticipated cloud rates.

An economic and energy-aware analysis of the viability of outsourcing cluster computing to a cloud

“…They also developed a web tool where the users can introduce the parameters of their Clouds and obtain the total cost analysis. Finally, other works such as [11], [12] or [13] have tried to compare the cost of owning a datacenter infrastructure versus the pay-per-use costs of Cloud deployments.…”

“…For the analysis we choose the average value of that scale, i.e. 2.15, although other cases might involve adjusting that ratio, such as in [13,18].…”

“…Therefore, all the internal nodes are powered off and only the frontend and the essential components remain powered on. The upper bound expression is shown in (13), and it corresponds to the theoretical state in which the cluster is working at full workload, where all the computing nodes are running and using 100% of all the computing cores (c). In (13), the amount of power consumed by the n nodes is estimated using the power of an idle node (P I ) and the additional power consumed by each of the c cores in the cluster (P C ).…”

Efficient and elastic management of computing infrastructures

“…They also developed a web tool where the users can introduce the parameters of their Clouds and obtain the total cost analysis. Finally, other works such as [14], [10] or [131] have tried to compare the cost of owning a datacenter infrastructure versus the pay-per-use costs of Cloud deployments.…”

“…For the analysis we choose the average value of that scale, i.e. 2.15, although other cases might involve adjusting that ratio, such as in [131,9].…”