This article is part of an Essential Guide, our editor-selected collection of our best articles, videos and other content on this topic. Explore more in this guide:
4. - Management, financial tools in your private cloud arsenal: Read more in this section
- Enhance private cloud benefits with automation
- Unlock private cloud's potential with automation, monitoring tools
- Introducing IT chargeback to meter private cloud use
Explore other sections in this guide:
- 1. - Private cloud 101
- 2. - Considerations for a private cloud migration
- 3. - Blueprints for moving from virtualization to the private cloud model
One issue that limits the adoption of private clouds within enterprises is that shared resources will be charged back to business units that use the private cloud. IT chargeback becomes a problem in companies where budgets are very tight and separate divisions don't work and play together as well as they should. Oh, you know … in most companies.
Fortunately, there are many third-party packages and services you can use to track your private cloud computing usage. These tools function as additional external applications that are loosely coupled to core applications and cloud services. Some exist on-premises and some are public cloud services that you subscribe to. While this technology goes by many names, the term I prefer is "use-based accounting."
The right approach and technology will be up to the requirements of the private cloud service and the business. In some instances, IT teams attempt to build accounting tools in house. However, that's usually a bad idea considering the technology is readily available to buy and somewhat proven -- as long as you select the right kind.
Once you've chosen your use-based accounting technology, you'll need to decide how you want to bill for IT services and resources. Most private cloud services will be billed using chargeback services within the same enterprise, leveraging internal dollars to pay the internal bill.
While your approach to internal accounting of private cloud usage may vary, I like to break them down into the following four general models:
- All you can eat
- Bill for time
- Bill for quantity
- Bill for instances
These four approaches mimic those of public cloud providers.
The "all-you-can-eat" approach tracks use of the service, but consumers (i.e., internal end users/business units) can use as much of the service as they like and pay a flat rate. This is applicable if you don't want to limit users from leveraging the service, and understand that maintaining a system to monitor detailed usage can be more trouble than it's worth since it does not typically require a use-based accounting tool.
The 'bill for instances' model is perhaps the best way to account for private cloud resources usage.
While this is the least sophisticated option, it's also the most popular. Companies that deploy private clouds estimate resource use by each internal entity. They define a number of dollars for those IT resources, and each month, money is removed from the entity's budget.
The downside of the all-you-can-eat approach is that while a few entities will make out well -- namely the primary users of the private cloud services -- the cost moves to entities that may have very little use for the private cloud, but are billed for it nonetheless. Kind of like gym memberships after everyone has given up on their New Year's resolutions.
"Bill for time," as you may expect, tracks the amount of time the IT service is in use and bills according to a set price for that time. The use-based accounting system will report the usage of IT resources and create the bill for you. Money is then removed from the budget based on the time consumers use private cloud computing services.
The upside of bill for time is that business units pay only for the time they use the private cloud service; therefore, utilization is better understood and billed according to use. However, time is not always a good indicator of resource use. For instance, while two entities are billed for one hour of private cloud service use, one entity may have saturated the private cloud while the other barely made the usage meter jump.
The "bill for quantity" model means IT pros bill consumers for the quantity of data that's transmitted to and from the private cloud, typically at the megabyte or gigabyte level. Again, the use of data is tracked, and a bill is automatically transmitted to the consumer of the private cloud service.
While this seems fairer than the bill for time model, in reality, data transmitted back and forth to the private cloud is not a great indicator of use of internal cloud resources. It's more of just the conversation with the requesting system or user.
The "bill for instances" model is perhaps the best way to account for private cloud resources usage. In short, IT pros watch the number of instances, such as storage and compute services, those who leverage the private cloud spin up and down. This gives you a more accurate view of who's doing what, with what, and therefore helps you determine how much they should be charged.
I suspect use-based accounting for private clouds will evolve a lot in the next few years, considering the growth of private clouds and shrinking IT budgets. Who knows, this may be a good way for IT to understand more about their end users.
About the author:
David (Dave) S. Linthicum is the CTO and founder of Blue Mountain Labs, an internationally recognized industry expert and thought leader, and the author and co-author of 13 books on computing, including the best-selling Enterprise Application Integration. Dave keynotes at many leading technology conferences on cloud computing, SOA, enterprise application integration and enterprise architecture.
His latest book is Cloud Computing and SOA Convergence in Your Enterprise, a Step-by-Step Guide. Dave's industry experienceincludes tenures as CTO and CEO of several successful software companies and upper-level management positions in Fortune 100 companies. In addition, he was an associate professor of computer science for eight years and continues to lecture at major technical colleges and universities, including the University of Virginia, Arizona State University and the University of Wisconsin.