This whole cloud thing. Does it work as a sustainable financial concept, or is there a tipping point that slides us back towatds a more traditional hosting model?
Firstly, we must recognise that not all clouds are the same, have different financial models and different purposes. This post doesn’t profess to be the fountain of all knowledge on cloud financial models, rather a musing on a potential for there to be an effectual glass ceiling or not where the model of clouds tips back in favour of a more traditional hosting architecture.
Long term average costs of productivity are assisted by a scale out and long term strategy – all heading in the direction of reducing the average short term costs by applying an economy of scale model to buying, managing and maintaining an IT infrastructure. Whilst managed hosting contracts tend to be fixed term (albeit often renewed), cloud models take the longer term to benefit from these economies of scale and strategy.
Cloud architectures are also by design pre-disposed to scale. As with a lot of things, generally the bigger the architecture and the better the cost saving associated with running the hosted service. Generally, it’s accepted that there is less of an overhead associated with running a cloud infrastructure due to the nature of cloud deployments – resilient architecture, live-live locations and failure-tolerant applications. However, to some extent this is also true about a traditional hosting environment. All the elements of capital and operational expenditure are present in both types of deployment – specifically:
- Volume purchasing (hardware / software).
- Hosting costs (local and co-location).
- Networking (local and commercial transit).
Other factors are found in the project costs associated with a cloud deployment that might affect the overall economy of the project. With a new platform, start-up costs would expect to be high for a fresh deployment, against significantly lower start-up costs for more traditional hosted environments that have already been deployed in a ‘cookie-cutter’ templated deployment. However, once running, the units of scale for cloud platforms decrease over time (as the infrastructure looks after itself through a self-service model and minimal administration), but the units of scale for managed hosting remains the same due to it needs to be constantly maintained by operational process.
Then real benefit to a cloud deployment comes in the operational cost (OpEx) of maintaining a cloud service. Deployed properly, OpEx of a cloud platform should be significantly less than a managed hosting environment. By using the self-service aspects of cloud operations, combined with the reduced overhead of maintaining a resilient platform designed with failure and uptime in mind – OpEx can only be lower than that of a traditional hosted environment.
Less administration required = Less manpower = Less operational cost = Lower project deliverable costs = Better profit margins!
Jeremy loves all things technology! Has been in IT for years, loves Macs (but doesn't preach to others about their virtues), loves virtualization (and does shout about it's virtues), and sometimes skis, bikes and directs amateur plays!