Taking the Guesswork out of ITIL

Where the “Rubber Meets the Road” in ITIL Adoption

So you’re adopting ITIL. Your goals are clear. You have ITIL-certified talent on your team. Maybe you’ve even begun evangelizing and vision re-engineering to persuade those who resist change. When you come this far, one undeniable fact soon becomes clear: drawing ITIL concepts on a PowerPoint slide is far easier than implementing them with thousands of servers and hundreds of legacy applications. This is immediately evident when you begin to tackle incident and problem management.

Truly effective operational incident and problem management require you to make sense of hundreds of thousands of system and application monitoring data points in real-time. Your ability to detect an incident is only as good as the alarms you receive from your monitoring infrastructure. To generate relevant alarms, you have to set accurate thresholds – usually manually – for the right set of key performance indicators (KPI) for every configuration item (CI) in your environment. Plus, you have to create dependency tree maps and write rules to cross-correlate the alarms that you receive to identify the root-cause across monitoring silos.

How do you do all of this for hundreds of thousands of data points across thousands of configuration items, under different workloads, for different times of the day, week and month? The dirty secret in incident and problem management implementation is that it is humanly impossible to meet this challenge using traditional monitoring and BSM tools.

Self-Learning Performance Management Emerges

If understanding the data points is humanly impossible, how do you accomplish it? A new generation self-learning performance management tool from Netuitive can help. Netuitive applies sophisticated multi-variable mathematical algorithms to analyze the data automatically and in real time.

In essence, the software begins by collecting KPI data from your existing monitoring tools – CPU utilization from the server, number of database locks from the database cluster, network traffic flow data from the routers, average seek time on the storage area network, etc. Then, it statistically cross-correlates the data to understand each KPI’s unique behavior pattern on each individual CI and its variations within the context of other inter-dependent metrics. The resulting thresholds are dynamic and calendar-sensitive, any deviations are considered in the context of other metrics, and trends are forecasted into the future.

Self-learning performance management technology eliminates the need to manually set thresholds, create dependency trees, or write down-stream suppression and other rules.

Fast Results: Accelerated ITIL Success, Two-month ROI

The results? Netuitive helps you accomplish your ITIL goals in a fraction of the time, with much less work, and with a level of accuracy that only mathematical algorithms and automation can deliver. The software deploys easily and offers rapid time to value. Typical return on investment (ROI) is two to three months. More importantly it enables the deployment of truly effective incident and problem management processes in days versus months or even years.

© Copyright 2007, itSMF USA All rights reserved. This article originally appeared in, and is reprinted with permission from, itSMF USA “The Forum”. The Forum is a bi-monthly newsletter published by itSMF USA.