Dick Benton, a Principal Consultant for GlassHouse Technologies, has worked with numerous Fortune 1000 clients in a wide range of industries to develop and execute business-aligned strategies for technology governance, cloud computing and disaster recovery.
Last month, we discussed the first of seven key tips IT departments should follow if they want to begin building a better service strategy for their internal users: knowing what you’ve got. That means ensuring that you can run a supply-and-demand model and inventory management on those resources you intend to deploy as an Internal Cloud Service Provider (ICSP). The next step, which we’ll explore today, is figuring out what it costs. You must develop a cost model so you can determine the cost per deployable unit of your compute and storage resources. You don’t have to have a charge-back, but you do need to be able to show costs and report on costed usage.
The very essence of services delivered by a public cloud provider (PCP) is that the consumer pays for the services they select. Amazon and similar organizations are not free and don’t/won’t "bulk bill" your company. The PCP bills the person who orders the service for the amount consumed.
Contrast this with how internal IT resources are delivered. The finance team is often loath to allow IT to cross-charge for services, typically because they see a lot of extra work for little perceived benefit and get a lot of push-back from profit center managers quite happy with the invisibility of their IT spend. Yet these are the same managers who will happily empower their development staff to use personal credit cards and the organization’s expense management system to acquire resources for rapid development and deployment of their favored applications.
IT seems to be between a rock and a hard place. When IT services are perceived as free, or at least available at no discernible cost that can be tied to usage, then costs will not inhibit usage and demand can increase without checks and balances normally associated with resource purchase. This is what we see in many organizations today: unfettered demand on limited resources, gold plated services deployed for mundane tasks and rising levels of dissatisfaction driven by perceptions of inadequate service and response. Too often IT is perceived as “always saying no” or putting up roadblocks. Captive consumers using internal IT typically see IT as an impediment, not an enabler.
Benefits of Cloud: Visibility
The success of your internal cloud deployment is dependent on your ability to understand the cost of the resource units you deploy and your ability to monitor and report their usage by consumers over time. Offering free services may be a great marketing approach (loss leader), but it is not a strategy for long term business success. Now you don’t need a formal “chargeback” process for this to be successful, although that is the preferable approach. Similar results can be achieved in what we call “showback.” By producing a regular report showing which consumers, departments or divisions are using what services at what costs, management attention can be focused on how appropriate this consumption may be in the bigger picture.
Is a low-value operation consuming high-value resources? Is a critical operation consuming the appropriate resources? Once resource consumption is subject to cost-conscious behaviors, we typically see a realignment of consumption and, equally, a realignment of services offered by IT as they can now deploy for demand rather than build gold-plated solutions just in case.
Build Your Own Cost Model
Figuring out the cost of your resources is a relatively simple process involving a spreadsheet and some basic data on purchase costs and depreciation, along with maintenance, facilities and staff costs. Building a suitable cost model will allow you to determine the cost of a configured GB of storage and the cost of a deployable “CPU,” including a memory cost. In this way, adding memory and CPUs to service-differing performance solutions becomes relatively simple. Standby resources for high availability also lend themselves to this calculation. Virtualization resources can be calculated through density ratios, and storage services may include options for backup, archiving and disaster recovery, each based on a per GB cost.
At step two of our seven steps, we now understand what we have to deploy (or sell), and we understand our own costs and what can be charged for the consumption of these services. Most importantly, we can tell who is consuming what and at what rates, supporting our supply and demand forecasting capabilities.
In our next post, we will discuss factors in building your own menu or service catalog.
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.