ABSTRACT
Metrics are di¨cult to present in a meaningful manner. What are you measuring? What does this metric mean? To the executive? To the sta«? How do you make people responsible for a metric? Does the metric include compliance or regulatory reporting? že bottom line is that metrics are meaningless unless they are tied to business outcomes. If the metric does not re¬ect a success or failure of the business in achieving a desired outcome, then there may be no point in reporting that metric. In e Visible Ops Handbook, from the Information Technology Processing Institute (ITPI) Behr, Kim, and Spa«ord (2004-2005) used Mean Time to Repair (MTTR) and Mean Time Between Failures (MTBF) as a metric that measures the uptime of systems. Measuring the uptime of systems is part of the metric. že real metric that should be measured is the impact on service delivery that should be measured. žis means the recovery of the service back to full production, not just the recovery of the server or application back into production. An outage of the infrastructure could be a router, server, or a workstation. It could be a virus or a denial-ofservice attack. že point is that until the business is back in full production, the impact is still being felt even if the infrastructure is back to full operations.
