Deploypartners Case Study
NBNCo – Network Assurance for a rapidly expanding network
NBN Co is rolling out an Australia-wide network to provide leading edge broadband service to all Australians. This network utilises three types of technology, depending on service location: fibre, fixed wireless and satellite.
Increased national demand for Internet service stems from new devices and applications that increase usage and require bandwidth growth. More and more people are using the Internet as a part of their daily lives – a year-on-year trend of approximately 50% annual growth in Internet traffic. The infrastructure project is a future-proof solution that will enable superfast broadband regardless of customer location, and ability to readily adopt future technologies.
The National Broadband Network (NBN) under construction establishes a Fibre to the Premises (FTTP) network with an operating life of several decades. Wireless and satellite networks to service locations where it is not feasible to deploy FTTP will complement the FTTP network. During 2012 construction of the NBN will increase from 24 to 49 towns and suburbs across Australia, with 570,000 premises currently having new service access. At completion the fibre network will connect 93% of Australian premises.
NBN Co require operational diagnostics which provide maximum understanding of the vast and rapidly expanding network. DeployPartners, working through IBM, were tasked to provide the service assurance tool for the Network Service Operations Centre (NSOC), & amalgamate all alarm information under a single platform. These alarms are sourced from a range of equipment that can provide operational diagnostics, with Tivoli Netcool™ as the top-level monitor of networks. Beyond this primary solution, NBN Co also required the capability to enrich network topology information to gain a better understanding of fault instances and their resolution.
DeployPartners designed and implemented a monitoring solution to meet NBN Co’s requirements, and performed this within a constantly changing project environment that demanded rapid response and high flexibility.
The most significant of these changes involved:
- ‘greenfields’ project, with no existing monitoring systems
- evolving the project requirements with changes to the project endpoint during implementation phases.
- having a high degree of organisational change, with teams forming and learning
- maintaining short delivery time, while changes to solution design needing quick turn around
- needing to integrate the solution with evolving customer processes (e.g. security policy)
External political influences also frequently impacted on the NBN project rollout, adding to the environmental fluidity.
Monitoring the Network
The DeployPartners Assurance alarm and event management platform uses customised capabilities based on IBM Tivoli Netcool™ product suite. The components were developed under a 3-phase design & release program throughout 2011 – the most recent release ‘going live’ in December 2011. This release incorporated the cumulative benefits of previous releases.
The most common fault occurrence causing alarm events is a break in fibre cable. The NBN Co network uses extensive cabling (in some instances travelling up to 60km long) with the potential for breaks to occur between exchange base equipment and customer end-points. Alarms can be initiated from either of these two points, and reported centrally.
The monitoring process comprises of the Event Management System which uses probes to search for raw events at several network domains.. As the order of cable branch points is known, clusters of contiguous service points that show fault alarms can be isolated to a single fault relating to a local cable. Network events captured from these areas of the network by Tivoli Netcool/Omnibus Probes, are processed to de-duplicate, enrich, and allow operator interaction. This sets up a truly ‘informed’ environment for fast event resolution. Operators are able to isolate faults as originating in specific domains, thereby quickly understanding the “trouble-to-resolve” and take the appropriate course of action.
Engine Room Components
Three Tivoli Netcool™ components formed the core of NBN Co’s ability to actively monitor the broadband network and enact quick fault repair:
IBM Tivoli Netcool OMNIbus: is a service level management (SLM) system that collects event information from many different network data sources and presents the data in a simplified view for operators. Collection of events from the NBN Co Network is via a series of probes with each probe being designed to receive events from a specific event source. OMNIbus provides first level event processing, aggregation and storage of all event information.
IBM Tivoli Netcool Impact: is an interface engine introduced to integrate incidents created through the network, ensure that duplicate incidents are not created.
IBM Tivoli Network Manager (ITNM): provides four key capabilities to the Service Assurance platform:
- Automated network discovery
- The ability to visualise the network topology
- Isolation of network faults via the Root Cause Analysis (RCA)
- Display network status information on Network View.
The NBN Co solution also incorporated High Availability functionality: probes using a peer peer failover system that offers a high level of availability. In this configuration, two probes run simultaneously. Should the master probe fail or shut down the ‘slave’ probe will no longer receive heartbeat messages (from the master probe) and the slave then begins to forward events. The result is continual real-time monitoring.
The capability to enrich service information was built into the 3rd phase of implementation. It is achieved through the Service Impact Analysis (SIA) tool, designed specifically for the NBN Co project. SIA utilises Impact & OMNIbus to further integrate and enrich instances after alarms are received. This presentation of event occurrences gives operators an even better understanding of how the entire network is performing.
- Centralised and highly specific monitoring and resolution of NBN Co network fault information
- Provision of a solution that can be modified in future to adjusting to a rapidly expanding network rollout
- Demonstration of a DeployPartners timely delivery in a changing and uncertain environment
- Advanced event processing and RCA providing identification of probable cause and fast initiation of trouble-to-resolve processe.
- Service Impact Analysis driven trouble-ticketing based on automatic identification of impacted services and customers
- Customer service centre has information about network outages and faults before customers call in to report problems