The most popular and comprehensive Open Source ECM platform
Let us know more
Economics of Cloud Computing: When Does Moving to the Cloud Make Sense?
Moving IT operations to the cloud saves money when compared to using on-site computing facilities. That’s a premise that’s increasingly believed, but the truth of which actually depends on a number of parameters. A group of academics from the Pennsylvania State University looked at the economics of cloud computing and concluded that cloud computing currently makes the most sense economically for small businesses and applications with small workloads.
The researchers divided costs into categories of being “quantifiable” and “less quantifiable”, and “direct” and indirect. Given this division, they constructed a matrix showing the classification of various costs.
| Direct | Indirect | |
| Quantifiable | Hardware (Server and Storage) Software (OS and Database) |
Shared Rack Storage Networking Infrastructure |
| Database and Operating System Maintenance | Staff Salaries | |
| Electricity used by Servers Cloud Usage Charges |
Cooling and Lighting costs | |
| Less Quantifiable | Software Porting | Security vulernabilities |
| Application Migration | Performance Charges | |
| Application Complexity | Time Delays |
The report considered the two option of pure in-house and pure cloud-based hosting, but it also considered combinations of the two approaches, which it termed as “hybrid” options. In the hybrid model, they identified vertical partitioning and horizontal partitioning. With vertical partitioning an application might be split so that part of it reside on-premise and another part of it resides in the cloud. Horizontal partitioning replicates some parts of an application in the cloud and, as load on the application increases, usage can burst or spill over onto servers running in the cloud.
The report found that ” (i) complete migration to today’s cloud is appealing only for small/stagnant businesses/organizations, (ii) vertical partitioning options are expensive due to high costs of data transfer, and (iii) horizontal partitioning options can offer the best of in-house and cloud deployment for certain applications.”
