Maximizing Operations and Security Effectiveness: Why Integrity Is Misunderstood, and Strategies to Cope and Thrive Despite It All

Apr 02, 2003

Abstract

While ITIL and BS17799 practices are emerging as the industry standard for
IT best practices, they do not describe in any usable way how to make actual
practices more resemble best practices. In other words, how do you
bootstrap good process where none already exists?

In this presentation, we will present the results of numerous benchmarking
excercises, where we found a most surprising correlation: the "best in
class" operations organizations with the lowest Mean Time to Repair (MTTR)
and the highest server/sysadmin ratios are consistently the most secure as
well.

To motivate why this is the case, we will present the Visible Ops
methodology, which is based on the ITIL concepts, and represents how "best
in class" operations conduct daily processes. The goals of the Visible Ops
methodology are to decrease Outage MTTR, improve operational efficiencies,
and build a "culture of causality" in operations. Visible Ops concentrates
on ensuring process integrity by using Tripwire in three service level
disciplines: Release Management, Controls, and Problem Management. Key
benefits of the methodology include a step-by-step cookbook for creating
processes, even where none currently exist, and a framework for generating
key metrics for continual process improvement.

This talk also describes the twenty years of computing history that has led
to the current state of affairs, including IT decentralization and
commoditization of technology, changes in the capacity expansion model
caused by the transition from mainframe to distributing computing, and the
continual tendency to misdiagnose the problems, addressing only symptoms
while the underlying disease remains misunderstood.

This talk is intended to provide practical IT audit and control
philosophies, in the context of classical IT defense philosophies, but
addressing the new problems posed by modern infrastructures.