|
 |
|
|||
|
|
|
||
|
|
|
|
|
|
|
|
|
|
|
Historically, business has abdicated the cost/risk tradeoff decision to the ITO, which (due to poor communication between the business units and the ITO) often led to infrastructure decisions that were not aligned with business strategy (for example, overspend to provide high availability or underspend on scalability or flexibility). In the new economic climate, ITOs are no longer able to simply "buy the best" to minimize risk (for example, time to solution and high availability) and instead must shop for a "good enough" solution that balances cost with risk, in alignment with the business strategy. This is particularly noticeable in storage, where rampant capacity growth (averaging 100 percent annually in distributed systems) has resulted in storage expenditure exceeding server expenditure, with storage now accounting for approximately 10 percent (and growing) of the total IT budget. We have observed ITOs attempting to compare wildly different technical solutions (such as Unix clusters with Linux/Windows 2000 farms for Web serving, and EMC storage arrays with DataCore virtualizations engines for storage) with different price structures in an attempt to lower infrastructure costs. When viewed as technology components and a technology decision, ITOs struggle with the selection process.
Where the technology is well proven (such as EMC storage arrays and Unix clusters), service levels can be robustly estimated from vendors' claims and validated against reference implementations and published analysis. In some instances, the service levels can be calculated from the underlying components. For unproven technology such as virtualization engines, ITOs must take a conservative view and estimate the service levels as those they can guarantee to fulfill. This means underestimating the service levels to ensure they can be reliably delivered. Once the service levels and unit costs are calculated, the technologies can be easily compared. Where a technology offers the same service levels at a lower cost, or higher service levels at the same cost, the decision is obvious and can be taken by the ITO. However, when a decision must be made to trade off higher service levels (lower risk) and lower costs, the ITO must involve the business. This is essential, because the consumer (the business units) must be held responsible for trading off risk for cost to align the service consumption with the business strategy. ITOs that attempt to make this decision will become a cost center instead of a service provider and frequently the scapegoat for the business's failures. Business Impact: To align IT infrastructure with business strategy, business units must control the infrastructure cost/risk/sourcing tradeoff instead of abdicating the decision to the IT organization. Bottom Line: IT organizations must define infrastructure as services to enable the business to select the IT infrastructure cost/risk/sourcing profile that ensures alignment with business strategy. Leading IT groups will present major new infrastructure investments as a portfolio of cost/risk/sourcing alternatives couched in business terms.
Bargain Basement Infrastructure: When Is Cheap Good Enough?
[an error occurred while processing this directive]
 [an error occurred while processing this directive] |
|
[an error occurred while processing this directive]
|
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
