Why you need to read up on the IT Infrastructure Library

By Mike Karp

With many enterprises now looking seriously at the benefits they
can squeeze from new technology trends such as grid computing,
information lifecycle management and on-demand (or utility)
computing environments, it is increasingly obvious that managing
change grows in importance. Why? Because at the heart of each of
these trends is the constant deployment and redeployment of
assets and processes to meet the changing needs of the users.
This becomes particularly challenging because in order for such
systems to run properly they must be able to look across - and
manage across - not just storage, but also most of the networks
and systems.

Managing, monitoring and analyzing such changes amid such
complexity are quite clearly impossible without an automated
infrastructure, but how can even the best set of automation
tools keep track of which asset is allocated where, what is
offline and what is online, and so forth?  The answer lies in
the concept of the Configuration Management Data Base (CMDB), a
centralized data repository that holds all information about the
status and changes in status for all IT assets.

The reality is that such a single CMDB will still be a long time
in coming, if indeed it ever arrives. However, a much more
practical concept than a monolithic CMDB is the idea of a set of
"federated" CMDBs, multiple databases that manage subsets of
changes and then share information with other CMDBs within the

Unfortunately, there is at present no accepted set of standards
that defines what the data within a CMDB or set of CMDBs should
look like. As a result, while many companies have developed
CMDBs to oversee the well-being of their own sets of products,
there is no way at present to pass data back and forth between
the databases and let them interoperate with one another as a
single consistent system.

The fact that such systems will eventually have to work together
in a federated fashion is clear; much less clear however is what
standards they will use to pass this information back and forth.
XML and CIM are two consistently recurring standards to which
many of the vendors outside of storage have turned; SMI-S from
the storage world should play very nicely with these.

At this point, the idea of a CMDB is increasingly appealing. But
how can we achieve the needed linkages between the various data
sets now in use? One possibility is ITIL.

ITIL, the abbreviation for the "IT Infrastructure Library," is
an acronym that almost never pops up in U.S. storage circles but
which our colleagues in systems and networking are beginning to
use. It is a set of standards originally promulgated by the U.K.
government, but that has now gained significant acceptance at
locations worldwide as the best guidelines to aid the
implementation of a framework for IT service management. By many
measurements, it has become the most widely accepted approach to
IT service management in the world. The ITIL standards appear in
a series of publications (hence, the "Library") that define best
practices for managing IT services.

ITIL is at present silent on the topic of standards that can be
used for CMDB interoperability. However, if IT managers
worldwide continue to subscribe to its ideas this may well turn
out to be the forum within which such standards are defined. At
the very least, it makes sense for U.S. managers to become
conversant with ITIL.  One good place to look is


Network World Radio: Getting started with ITIL

IT processes ranked
Network World Fusion, 03/30/05

Tech Update: SMI-S 1.1 simplifies storage management
Network World, 06/06/05

Copyright Network World, Inc., 2005

Questions or problems regarding this web site should be directed to abeckman@outdoorssite.com.

Copyright 2008 Art Beckman. All rights reserved.

Last Modified: March 9, 2008