The Obama administration made cloud computing an integral part of its 25-point plan to reform federal IT. On Dec. 9, 2010 agencies were mandated to identify three projects that could move to the cloud within the next 18 months. But, how can an agency chief information officer know if a move to the cloud will be more economical than in-house hosting of that service or program?
A group of three researchers from The Pennsylvania State University set out to answer that question in their paper, “To move or not to move – The economics of cloud computing.”
They concluded that cloud computing is most appealing for small businesses or workloads.
“For small workloads, the servers procured for in-house provisioning end up having significantly more capacity than needed (and they remain under-utilized) since they are the lowest granularity servers available in market today.”
As for larger workloads, the researchers said, “For medium workload intensity, cloud-based options are cost effective only if the application needs to be supported for 2-3 years, and become expensive for longer-lasting scenarios.”
The researchers estimated costs for cloud-based projects, non-cloud projects, and hybrid projects that combined both, taking into account how costs for these projects would grow over time. The researchers looked at what they called “quantifiable” costs such as hardware, software, electricity consumed by servers, cloud usage charges, and staff salaries, and left out “less quantifiable” costs such as application migration efforts, security vulnerabilities, and software porting efforts.
Researchers Byung Chul Tak, Bhuvan Urgaonkar, and Anand Sivasubramainam presented their research at a recent conference. According to PC World, the researchers were criticized by one audience member who said the less quantifiable costs needed to be considered in order to have a more accurate picture of the costs associated with both cloud migration and traditional in-house options.