This is a great challenge for both you and your clients. Essentially you have to do more with less. Every environment is different, but in every environment there will be a group of events that cause the majority of the problems and threaten to prevent a data center from meeting service-level requirements. If you're able to catalog the events and address them, you and your client will be able to meet the service-level requirements.
For example, the majority of these events may be generated by hardware problems. Or they may be generated by user errors. Analysis of the root causes will help you address the problems.
In some cases the service-level requirements and the budget cuts may be unreasonable and your client will have to make a well-documented case for being unable to meet them, and then negotiate for a new, less stringent service-level requirement or an increase in budget. Until your client is able to negotiate increased budgets or less-stringent service-level requirements, they will have to ensure that their team is able to meet expectations without a catastrophic failure or staff turnover.
This was first published in December 2007