Availability Metric for IP-Based Networks and Services
|Publication Date:||1 October 2008|
SCOPE & PURPOSE
Service availability is a critical component in designing IP-based networks. It is also a key measure in Service Level Agreements (SLA) between service providers and customers. Hence, it is important that the concept of availability is well-defined and understood, and that the capabilities for accurate estimation are readily available and implemented by service providers. This document provides a definition of a commonly accepted availability metric for IP-based networks and services that can be correlated to network element outages. Future documents will discuss this metric within the context of MPLS-based transport networks.
Current network trends indicate a significant increase in traffic of different types over IP networks. Traffic types include real-time Voice over IP (VoIP) and video services, data services, "Enterprise Customer" services over "Virtual Private Networks" (VPN), IPTV services, as well as traditional Internet services such as e-mail and Web browsing. In designing SLAs with critical customers for key services such as VPN and video, service providers typically agree to a specified level of availability within the bounds of a pre-determined traffic volume for the service in question. Under such conditions, the predominant and overwhelming contribution to the unavailability of the specified service is a failure or outage in the network. A hardware or software outage may result in downtime for one or more network elements, thus resulting in service disruption with negative impacts on service availability.
This standard describes a metric that quantifies the impact on service availability of an underlying network element outage. Currently, Network Management System (NMS) tools offer limited capabilities to collect necessary data for estimating this metric. The purpose of this metric is to encourage development of outage measurement capabilities/techniq
The scope of this document is restricted to estimating service availability impacts within a single IP network domain. It is also restricted to service availability impacts caused by network element outages; it does not include service unavailability periods outside of such conditions (e.g., service degradation as a result of packet loss or jitter). It is recognized that communications services can span several network domains, possibly of differing technologies (e.g., 3GPP Access-IP Backbone-PSTN Termination). Once agreements have been reached on service availability metrics and estimation methodologies, discussions on extending availability on an end-to-end basis over all domains and technologies can commence. Finally, it is recognized that evolving technologies such as MPLS are further transforming IP-based networks. This document serves as the first in a series of documents that will also examine availability estimation within the context of MPLS-based networks and services.