The Binomial Bookstore

MANAGING SERVICE LEVEL QUALITY ACROSS WIRELESS & FIXED
NETWORKS
by Peter Massam

“If you need to know the implications of running multimedia applications when
traversing from
fixed to wireless networks and the guaranteed levels of service they require, then
this is the
book for you ..

“MANAGING SERVICE LEVEL QUALITY ACROSS WIRELESS AND FIXED
NETWORKS
deals with the search for the real time information that constitutes the 'customer
experience'
in terms of application performance so that service levels can be verified against
measurable
and relevant data in a true end-to-end manner across both fixed and wireless
networks.

“Examining what mechanisms already existed in fixed IP data networks prior to the
introduction of probe and agent technology this volume then looks at these later
developments using a real-world scenario of how real time application performance
monitoring can not only provide service level management but also aid in root
cause
analysis.

“This same model is then applied to a wireless environment, examining what
elements are
required to be able to deliver multimedia services across 2G and 3G mobile
networks,
detailing the components of data networking that will assist in guaranteeing service
level
performance and the constraints placed on those guarantees when passing
services over an
air interface to a wireless-enabled device.

“It tracks QoS components and mechanisms of both fixed and wireless
environments and
looks at how they will provide the glue in this brave new converged world.

“Written for IP professionals from either a wireless or an IP data networking
background
Managing Service Level Quality across Wireless and Fixed Networks offers
innovative
sound
and practical advice. It will also provide a complete reference guide for academic
staff and
students with an interest in this area.”

- - - - - - - -

EXCERPT FROM THE INTRODUCTION

“In 1998, as part of a global deployment of a mission-critical monster of an
application, I was
tasked with a brief to guarantee that the mythical source of all performance-related
problems,
which has come to be known as `the network', was not to be blamed for
degradations
experienced by the users of that application. At least, if it were to blame, we
should be able
to
produce the hard evidence to substantiate the claims and provide corrective
actions quickly
and effectively.

“On the face of it, this would appear to be a reasonable and simple request, but
isolating the
network portion from the rest of the transaction does not give the full picture.
Comforting
though it may be to know that your part of the network is functioning correctly, it
does nothing
to identify the source of performance problems, but rather remains a source of
constant
irritation to business and consumer customers alike. Like the Emily Pankhursts of
their day,
customers of these applications have risen up to demand what was rightfully theirs
- a decent
level of service.

“Embodied in that right are many different interpretations. For some customers, it
may be
working at the speed you and your thought processes want to, for others, it would
be working
anywhere at any time at the speed you want to. In both cases, this really boils
down to how
responsive a function or an application is. Put more simply, when you press the
button or key,
how long do you have to wait to get a response.

“It was clear that a solution had to be found that encompassed all portions of a
transaction
from key press through the network to the server and back again, and that
slowdowns and
their sources were flagged up at the earliest opportunity before customers were
impacted. In
the fight of this, a service-level management strategy was devised, planned, and
implemented globally within 12 months.

“It is that experience that I wished to share, as it became obvious that not only
was there a
good deal of interest from both internal and external customers in that strategy, but
also, one
of the fringe benefits, namely truly proactive end-to-end management, was being
realised.
That account with an example of the practical implementation of such a strategy
occupies the
first half of this book.

“The second half reflects both the movement in the industry towards mobile
computing and
the particular challenges when trying to apply a similar strategy across fixed and
wireless
networks. Here, the approach is one of familiarising yourself with wireless
environments,
understanding what is going to help you maintain service levels, and providing
some
indicators as to the expectations you need to set when passing what have become
familiar
applications over the air to a mobile device.

“To understand these challenges, it is important to understand the technologies,
what lies
behind them, and what mechanisms are available to assist you in capturing the
all-important
real-time information on application performance, which will convey to you the pain
as well as
the pleasure that your customers are feeling.

“Both halves cover aspects of technologies that relate only to application
performance. It is
not meant to be a thorough examination of SNMP, M113 structures, or RMON
groups, which
have been adequately covered by many acknowledged in the references section.
What it
does give is an insight into why service levels are important, how to implement a
service-level
management strategy based on application performance and what performance is
likely to
be like in a wireless environment.

“The test results in the concluding section of this book are meant to promote
further
investigation by those interested in the area of application performance.

“This was written both for IT professionals, from either a wireless or an IP data
networking
background who wish to familiarise themselves with the other half of the equation,
and for
anyone with an interest in how service quality is maintained and delivered across
different
types of networks.

“Armed with the information contained herein, I hope that you will feel better
equipped to deal
with the challenge that bringing fixed and wireless worlds together presents and
that it sets
you well on the road to delivering a service quality that your customers will
appreciate and
that you can be proud of.”

- - - - - - - -

EXCERPT FROM THE INTRODUCTION

“Understanding applications' performance is to know the frustrations felt every day
by
business and consumer customers alike. Without it, we divorce ourselves from the
reality of
slowdowns or ‘performance brownouts,’ which undoubtedly cost businesses a tidy
sum in
lost
time and inefficiencies. For those whose job it is to look after such customers, it
can be a
lifeline to find a means of gleaning real-time application response time information
to aid in
root-cause analysis before a business-impacting event takes place.

“Added to this, there is an increasing expectation that not only must the network
deliver on its
99.999% availability promises, but that applications' response time should also be
an
intrinsic part of any Service Level Agreement (SLA) between service provider and
customer.

“In recent times, the technology industry has been promoting the ‘anytime,
anywhere’ concept
of the all-pervasive Internet by talking up the ability to deliver all services to any
handheld
Internet Protocol (IP) device across the airwaves and is now being gripped by the
grim reality
of having to invest heavily for the privilege. Much of the fervour in 2000 centred
around the
auction of licences for the third-generation (3G) mobile networks and the
preparation of
applications to run over them, but we also saw the relaxing of restrictions in
summer 2001
allowing access to wireless Local Area Networks (LANs) in public places, which
has done
much to bolster a flagging LAN switch market desperately looking for new products
to justify
new price tariffs.

“While vendors vie for position over the ultimate access device, developers and
Internet
Service Providers (ISPs) promote their portal as the best way into the Internet, and
standards
bodies endeavour to keep up, it is easy to lose sight of a basic prerequisite to this
activity:
will it work?

“We begin with an introduction to real-time applications performance as a means of
delivering service level management (SLM) and of aiding root-cause analysis in
fixed
networks. We then go on to examine what it takes to guarantee service-level
quality across
fixed and wireless networks today, what is on offer tomorrow, what architectures
have to be
accommodated in this new, dare we say, converged model, and what levels of
performance
we can expect from it.

“As with many new concepts, terminology can be misleading. So, for the purposes
of this
book, the term 'multiservice networks' simply alludes to networks that carry
multiple services
namely voice, data, and video.”

- - - - - - - -

TABLE OF CONTENTS

- Introduction
- Managing Service Level Quality in Fixed Networks
- History
- Current Standards
- Latest Developments
- A Model Agent
- An Integrated Model
- Application Performance in a Nutshell
- Service Level Quality across Fixed and Wireless Networks
- Wireless Architectures
- Moving to Wireless
- Wireless Performance
- Conclusions
- References
- Appendices.

- - - - - - - -

ABOUT THE AUTHOR

“PETER MASSAM has more than 18 years IT experience spanning mobile, carrier,
enterprise, retail, government and education markets, primarily in the area of
end-to-end
management of voice and data networks, including 3G wireless multi-service
networks.

“At 3 (Hutchison 3G Ltd) Peter headed up the Service Assurance function, creating
a strong
foundation of service quality standards across 110 suppliers, introducing customer
and
network quality improvement programs, producing a complete ROI business case
for a
Business Markets product based on customer experience monitoring, and
delivering
comprehensive board-level business metrics. Prior to that, he was responsible for
the
strategy, design and implementation of a unique global Service Level Management
solution
at Nortel Networks, which led to the development of a new managed service.
Having
previously managed technologies ranging from Help Desk to back-end servers, he
has
developed extensive expertise in business, customer and operational service
management
requirements.

“Peter has also published a paper entitled “Decent Service Levels,” IEE
Communications
Engineer, June 2003 that illustrated an innovative method of maintaining service
quality
across multiple operator domains using today’s technologies.

“Peter is an Arts graduate from Durham University and received a distinction in his
MSc in
Professional Computing from Staffordshire University.”

- - - - - - - -
2002, 216 pages. Order #DR669
- - - - - - - -

The Binomial Bookstore

Service Levels & Service Quality

Rothstein Associates Inc.