QoS: When Service Assurance Matters…

In a modern business climate, where competitiveness, cost control,
complexities and return-on-investment are central to business processes,
assuring the highest service levels in operations and maintenance is critical
for service providers. Meeting these goals allows them to gain and retain
customers while differentiating their services and guaranteeing services leading
to more profitable customers.

On the other hand, enterprises are constantly trimming operational
expenditure on communications services and IT infrastructure, compounded by
constant worries about quality of service (QoS) offered by service providers.
Enterprise CIOs are on the lookout for efficient utilization and improved
performance of their internal communication systems, which contributes to the
bottom-line.

SLAs
and Objectives

The service level agreement (SLA), a contractual agreement, plays a key role
in solidifying the relationship between service providers and enterprises. For
service providers, an SLA is part of an assurance strategy that applies to
network devices as well as to customer retention and billing practices. For
enterprises, an SLA describes the most quantifiable means of resolving internal
billing worries. The point at which these SLAs converge, describes an optimal
performance management scenario, where service providers maximize the
infrastructure RoI and enterprises minimize communications expenses.

Service quality management is fast becoming an area for development and
implementation in these organizations. The aim of such solutions is to automate
the process of implementing, measuring, computing and reporting SLAs, both in
real time and historically. The need for SLAs extends across all types of
services from simple leased line and voice subscriber services to complex
value-added services. SLA metrics could be network uptime, throughput, mean time
to repair, billing accuracy and more.

In other words, the software needs to perform the function of a neutral and
trusted ‘arbiter’, with built-in ability to program the ‘arbitration rules’,
which are then monitored for compliance and resolution.

In these contexts, some very important questions worth answering are:

n How do we measure an end-to-end
service level?

n How do we provide instant
information on service levels being complied with on a per customer basis?

n How do enterprises track
through a single window service level compliance of external vendors and
internal IT organizations?

Total Service Level Management

Maintaining a sophisticated end-to-end service level management
requires a new approach based on an integrated operations support system (OSS).
Traditional fault management systems are inappropriate for this application as
they cannot process, calculate or report on service metrics. Likewise,
network-based performance systems are insufficient due to the fact that
customer-perceived service level metrics can be quite different from network
performance metrics.

At the heart of the service-centric integrated OSS lies total service level
management. This application processes and stores all the information relevant
to monitoring service levels from all aspects of the service, regardless of
whether the source is an IT or network infrastructure. Further, it has advanced
integration capabilities to link to other applications such as order handling,
CRM, trouble ticketing, billing, and workflow. The focus is on a true service
view, distinct from the network infrastructure, which integrates all of this
information, distributes and computes the critical service level metrics
according to defined business rules. Thus, this application forms the pivotal
SLA monitoring point.

Service-centric OSS

A service-centric integrated OSS would combine the SLM functionality with
new network and service management solutions that handle diverse range of
technologies, platforms and services. It would also include scalable
applications for fault management, performance management, network configuration
and activation. These applications should share, as much as possible, a common
communications infrastructure for increased performance.

It should be a conscious endeavor of service providers to achieve an
integrated view to monitor and manage networks, services and related SLAs. This
service-centric OSS should be flexible enough to address a smaller set of
requirements, with the ability to scale up as requirements multiply. The open,
modular design also enables organizations to take advantage of a variety of
third-party software.

John Mathew, Suresh Vaidyanathan, and Bithin Talukdar Hewlett-Packard