Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

A centralised management information system for availability measurement and reporting in mobile IP networks including GPRS

IP.com Disclosure Number: IPCOM000007513D
Original Publication Date: 2002-Apr-02
Included in the Prior Art Database: 2002-Apr-02
Document File: 5 page(s) / 136K

Publishing Venue

Motorola

Related People

Eoin OSullivan: AUTHOR [+2]

Abstract

This paper provides a description of an availability measurement system for IP networks with the emphasis on GPRS.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 32% of the total text.

A centralised management information system for availability measurement and reporting in mobile IP networks including GPRS

Eoin O’Sullivan, Chryssa Dislis

This paper provides a description of an availability measurement system for IP networks with the emphasis on GPRS.

Introduction

This paper relates to IP mobile network technology, specifically GPRS (General Packet Radio System). It describes an availability measurement solution for IP networks, with the initial emphasis on GPRS. The technique described enables close measurement of system availability, with inherent handling of the network topology and dependency relationships. The availability measurement is valuable additional functionality for operators, by enabling trend analysis of outages, linking to statistics to identify revenue losses and allowing operators to undertake pre-emptive solutions to expected problems.

Definitions

Availability = (uptime during reporting interval / total reporting interval)

Outage: period of downtime, i.e. the time a node or process fails to operate

Problem(s) To Be Solved

In a GPRS system, a number of alarms are used to inform the management system of network behaviour. These have different levels of severity from Low to Critical, but outage events are not explicitly identified. The identification of outage events is critical, as these events impact the operation of the network and ultimately affect revenue. In addition, the lack of outage identification means that there is no way of examining historical failure trends. There is currently no design way around this problem, other than providing the additional functionality described here.

Proposed Solution to the Problem(s)

The proposed solution identifies network and node outages for all network nodes (IP based), translates an event combination into an outage with a specified start/end and provides a method of examining failure trends in GPRS networks. Failure data is presented and reported in a centralised system, thus enabling historical analysis of failures. Outage trends can be examined in detail and pre-emptive solutions to expected problems can be put in place. This system is also ideally placed to compare software releases, third party equipment and third party services in respect of their impact to the stability and availability of the network, and can form part of regression test on a reference system

The proposed operation of the system is as follows. Numbering refers to the diagram numbers in Figure 1.

1.       The data source is the SQL database of configuration and alarm data for the GPRS network. This resides on the server.

2.       Data extract: Only a subset of the data set is needed for the availability measurement system. Specifically, the minimum set is the node name, node type, network configuration and the outage trigger set.  The trigger set is a subset of the alarm data which allows a node outage to be identified.

§         Typically an outage is specified by a pair of triggers, defining the beginning...