Introduction
IT infrastructure consisting of networks, servers, databases, and even
parts of application systems forms a networked computing system (NCS)
whose performance must be actively managed to ensure continual business
support. But the skills and tools necessary to ensure that network and
server systems provide adequate levels of services and performance are
expensive and scarce.
Unfortunately,
NCS performance management is frequently broken into a number of processes
that are even more frequently well intentioned but nonetheless neglected.
More often than not, an IT organization will have some form of network
monitoring system, but few have network performance measurement tools.
Similarly, relational database management systems (RDBMSs), server operating
systems, and application system performance management is usually conducted
as a part-time (often reactive) activity of the associated technology
administrator. However, management service providers (MSPs) that specialize
in performance management can apply expert personnel and 24/7 monitoring
at a fraction of the cost required to staff the function internally.
Note:
This note first appeared in a column by James F. Dowling in Mid-Range
Computing. Look for other previously published Mid-Range
Computing columns by Mr. Dowling at this site or visit Midrange
Showcase at www.midrangecomputing.com/showcase/.
What's
Involved?
A comprehensive networked computing performance management program encompasses
the following:
- Architecture
to establish a plan for how NCS components are interrelated
- Service
level commitments to link NCS and business performance measures
- Availability
measures to record uptime
- Activity
measures to record resource consumption
- Performance
measures to record quality of service
- Event
management and response escalation to reduce downtime/poor performance
time
- Fault
isolation to pinpoint performance fault cause
- Capacity
measures to record resource availability
- Capacity
modeling to plan for resource expansion
- Data
integration and reporting to bring various infrastructure component
data into a cohesive, business-relevant picture of the NCS's historical,
current, and future capacity to support business needs
NCS performance
management systems are composed of knowledgeable technology administrators
- who understand each of the NCS components well enough to prevent, identify,
diagnose, and plan resource performance - and tools to assist them. Although
performance measurement and event management tools are costly, the human
resource costs make comprehensive programs prohibitive for many. This
is not to say that such systems are not cost-effective. Rather, cost justification
is extremely subjective, based on projected faults and business loss estimates.
When coupled with high capital and human resource expenditures, such ROI
presentations are not well received unless a recent catastrophe can be
used as a case in point.
That's where
the MSP comes into the picture. Entuity, InteQ Corporation,
iSharp, Luminate, and Manage.Com are founding members
of the MSP Association, a consortium of companies that will define and
promote the emerging management service provider sector. Information on
the MSP Association is available on the Web at www.mspassociation.org;
from Association headquarters at 401 Edgewater Place, Suite 500, Wakefield,
MA 01880; via telephone at 781-876-8830; or by sending email to info@mspassociation.org.
MSPs bring
component expertise, measurement, event management, and reporting capabilities
to bear on the networked computing systems of multiple companies through
a single infrastructure investment. This allows MSPs to provide 24/7 monitoring
and event management coverage with technology component experts for analysis
and planning. Clearly, this is a program that every IT manager wants but
one that most cannot afford. The principle advantage of the MSP approach
is that a company has access to performance management experts and integrated
monitoring at a fraction of the cost of in-house staffing.
Two
Players' Approaches
The MSP approach is a new business model, and every situation presents
unique challenges to designers of monitoring and measuring systems. However,
two companies have made significant headway in this marketplace. I present
their offerings here as representative examples of what is to come and
what every IT manager must consider.
Luminate
(www.luminate.com) takes SAP R/3 performance management from the level
of drudgery and mysticism to that of fact-based decision-making. The company's
operating model is simple. The client downloads and executes a free monitoring
module for SAP R/3 and authorizes it to commence data collection and transfer
to Luminate's Intelligence Center. From this point on, serious performance
problems will be routed back to the client via email with appropriate
response advice provided. For proactive reporting, capacity planning,
and other analytical services, the client contracts an appropriate service
package. Other monitoring modules are available that cover a broad spectrum
of infrastructure components.
It
is easy to discount the value of being told that everything is all right
99.999 percent of the time. But it is hard to overestimate the value of
having trend data collected, analyzed, and reported without having to
think about it or being able to reassign an SAP R/3 BASIS System Administrator
to an activity with more apparent business value contribution without
losing management continuity
InteQ
(www.inteqnet.com) presents an operating model that mirrors Luminate's
simplicity while broadening services. InteQ provides a complete outsourced
infrastructure for monitoring, measurement, and event management of complex
networked computing systems. Instrumentation, using commercially available
tools such as Hewlett-Packard's OpenView, BMC Software's
PATROL, and Remedy Corporation's ARS among others, is networked
to InteQ's Network Operations Center, where it is monitored in accordance
with predefined service level agreements. Inteq's consultants can deliver
a full IT infrastructure life cycle from architecture design to monitoring
and management in an effort to ensure that the infrastructure is well
engineered, measured, and managed.
These
two companies, along with their partners in the MSP Association, are making
what has been possible yet virtually unobtainable for quite some time
into a turnkey service that brings many IT managers' vision to reality.
This
column will continue to explore the change/size paradox-big companies
desiring speed and growing companies desiring stability. The author would
appreciate feedback on material presented as well as suggestions for future
study and reporting. The general theme is IT management and the goal is
to make it easier to get clients what they want and what they need to
succeed.
About
The Author
Jim Dowling is VP of the Alignment Consulting Practice at TechnologyEvaluation.Com,
Inc. located in Woburn, Massachusetts. TEC researches IT products and
suppliers as well as the ways companies obtain business value from IT.
TEC's consulting services remove time, risk and ultimately cost from IT
related decisions.
Jim
can be reached at jdowling@TechnologyEvaluation.COM.