Real money versus assumed risk profiling is the most apprehensible part of the discussion around cloud economics. Advocates of cloud computing will have difficulties in making a business case for the cloud if they do not have clarity on how much real money they will be saving for the organization by migrating to the cloud.
The assumption is that cloud computing is cheaper. The reality is– it is cheaper! But, how does one prove that? There are direct and indirect costs of running a data center and problems associated with power, floor space, storage, IT operations, network and scalability issues. These need to be highlighted and evaluated in monetary terms. On the contrary, it must be underscored that cloud-computing services are offered at a nominal fee per GB per month, with none of the data center costs detailed above. The only other associated cost is bandwidth usage.
Definitions of appropriate data for cloud storage can be used to build a case for real money savings in the cloud. Not all data are candidates for latency in the cloud, some tiers of data are appropriate for cloud storage and others not. An effective business case for tiered cloud and offsite storage can be made in monetary terms by answering the following questions:
- What are the performance requirements for the category of data to be stored?
- What are the performance tradeoffs of using one platform over the other in terms of platform construction, application, deployment, compute rates and completion times?
- What are the specific resource requirements and monetary costs of creating and deploying applications in each platform?
- Can cloud-computing platforms be used in conjunction with local offsite systems to improve cost effectiveness?
Security and compliance are automatic in the cloud and legal discovery is facilitated. So, one cannot ignore the real money savings from avoidance of litigation and enhanced compliance to legal mandates.
Having said all this, it is necessary to point out that cloud services and infrastructures need constant monitoring. While cloud services can help enterprises up-front costs for hardware and software by spreading such expenditures over months and years, CIOs need to remain vigilant as costs creeps can occur in pay as you go pricing services. If data that is no longer required in the stores is not removed or destroyed from storage to free space, they will find it impossible to keep down costs of maintain the storage in the cloud at economical rates.