|Oracle® Communications Network Intelligence Concepts
|PDF · Mobi · ePub|
This chapter provides an overview of Oracle Communications Network Intelligence.
Network Intelligence is a capacity planning system used by Communication Service Providers (CSPs) for communications networks. It reads data from your network inventory systems or other network data sources and processes that data to create useful views of your network data. Network Intelligence allows network planners to view network reports and maps, plan network build, and identify potential capacity stress points within the network in advance. After loading inventory data in Network Intelligence, you manage your network assets, forecast traffic demand, plan network capacity, optimize network traffic, and perform network cost reduction and consolidation.
To use Network Intelligence, you load your network inventory data into the Network Intelligence database. Network Intelligence uses that data to provide a variety of views and reports.
You can load inventory data whenever you require, and it can also be done on a predefined schedule. When you first deploy Network Intelligence, you load all the data. Thereafter, you typically run automated loads at regular intervals (for example, every night during network downtime) that load only new data or changes to existing data, such as updates or deletions.
Figure 1-1 shows the Network Intelligence data sources including inventory management system (IMS) data, asset register information, and supplier list price data used in planning calculations. Network Intelligence can load and process cost data for network components that you can use to create costing plans for proposed trail routing.Within Network Intelligence, the data capture platform is displayed and the principal functional areas are listed, including capacity forecasting and trail routing. Data outputs include capacity planning, financial modeling, and reporting modules.
The Network Intelligence core platform shown in Figure 1-1 consists of the basic application platform and the data modeling database. The core module provides visualization, utilization, and trending reporting on all network entity types from topologies, networks, equipment, equipment holders, and links, down to individual cards and trails. By measuring and reporting the utilization of all network elements, and examining existing capacity utilization black spots, the core module suggests the re-routing of inefficient trails, and generates comprehensive outage information.
See "Displaying Network Data" for more information.
The Network Intelligence data capture platform consists of a core database and a staging database with corresponding (though separate and distinct) schemas as shown in Figure 1-2.
You can view multiple networks, with data spread across multiple inventory management systems (IMS), as a single network in a federated network view.
Custom data loaders extract data from an IMS. This data is cleansed and normalized before bring written into the Network Intelligence staging database.
The staging database is a storage area that provides continuous access to application data. The staging schema describes the layout of the staging database and consists of a subset of tables based on the equivalent tables in the core schema. These tables are generally equivalent to those in the core schema with three additional DELTA columns to identify newly created records, updated records, and deleted records. For example, the SITE_STG table contains all columns in the SITE table with three additional DELTA columns.
The core database is a storage area created to model and manage application data, typically from an IMS, or multiple inventories. It is the production database for Network Intelligence, and uses a core schema to describe the way in which the application data is organized into tables.
Each customer implementation has different data content from various data sources that requires mapping and synchronization to the network model depicted in the database tables. The core data loader extracts the data from each table in the staging schema and loads it into the equivalent table in the core schema.
For further information on schema tables, see Network Intelligence Developer's Guide.
For further information on database schema creation and configuration, refer to Oracle Database installation documentation.
Data capture takes place in three phases:
You use a custom loader to connect to the your IMS, retrieve network inventory data, and place it in staging schema database tables. The custom loader is generally written in PL/SQL or Java, but the data loading design and implementation may be accomplished using other tools (for example, Oracle Data Integrator).
Data is extracted from the staging schema tables and loaded it into the equivalent tables in the core schema. The data loading process is usually scheduled to run every night to ensure that the data presented in Network Intelligence is accurate and up to date.
Data is converted using a custom data loader into the format expected by the Network Intelligence data model. This process ensures that tables or entities represented in the IMS data model are correctly translated to the Network Intelligence data model. As every data load from an IMS at a customer deployment is unique, a custom data loader is generally required to be written for each implementation of Network Intelligence. The loader can be written by the customer, by a systems integrator, or by Oracle Consulting. A custom data loader consists of PL/SQL procedures that run against the input data. Transformation is part of the data conversion process; once the SQL script for a data loader is triggered, extraction, loading, and transformation take place seamlessly. Defaults can be created for any data required by Network Intelligence which does not exist in the IMS.
Loading data consists of synchronizing the core schema tables (the Network Intelligence database) with any changes that have occurred in the staging schema (the inventory data) overnight. This loading process is performed by PL/SQL procedures.
Network Intelligence can be configured to load data incrementally or in bulk. When configured to load data incrementally, Network Intelligence verifies the staging tables for new entries, deletions, and modifications, uploading the changes only since the last upload to the core schema. When configured to load data in bulk, Network Intelligence uploads all the data from the staging tables, adding it to the core schema.
When loading data in bulk, Network Intelligence uploads the data much faster and more efficiently than when loading data incrementally, by ignoring constraints and indexing.
You can configure a defined schedule for data loading, or you can trigger Network Intelligence to load data on demand. Oracle recommends that you load data during system inactivity, to minimize disruptions to system performance.
See Network Intelligence System Administrator's Guide for more information about configuring how and when Network Intelligence loads data.
See Network Intelligence Developer's Guide for information about creating and implementing the data loader.
The network can be viewed in different ways (for example, through topology, location, equipment, trail, transport, service, or customer perspectives). Figure 1-3 shows constituent trails grouped in folders by technology type.
See "About the User Interface" for more information about the features of the user interface including the tree browser, toolbars, and entity views.
Network elements (NEs) are active elements within a network subsystem used to provide a telecommunications service (for example, a network, topology, site, equipment, or trail).
Network elements in Network Intelligence are generated in one of the following ways:
Created in the Network Intelligence GUI.
Loaded from the inventory management system (IMS) in which they are modeled.
See "About Data Capture".
The tree browser displays a navigable list of all NEs in the Network Intelligence database (for example, the equipment tree browser displays all equipment currently modeled in Network Intelligence ordered in folders denoting equipment type, equipment sub-type, and definition).
Double-click a folder to expand the selection. Sub-entities within folders are populated on selection. Click any network to display a network detail view.
Each NE type has several views, menus and sub-menus available. For example, a network has the following views available:
Network Node view: Displays a logical view of the network's equipment nodes and their connecting trails.
Equipment Search view: Displays a filterable table view of the network's equipment.
Child Trails view: Displays a filterable list of trails that reference nodes of the network.
Customer Bandwidth view: Provides a complete list of every customer who is subscribed to a service carried on the network.
Determining the lowest-cost, end-to-end, trail solution between two sites in the network is critical to network planning, service fulfillment, and service assurance.
To find an optimal route for a new service demand in Network Intelligence, enter the start and end points of the service and data about its use, such as the customer using it and the capacity required. Using this data, a list of possible path routing solutions is generated, and the costs involved in the path provisioning process are calculated.
You can configure constraint rules to apply conditions when looking for path solutions for a service demand. You can also select policies that use routing rules to apply conditions on particular entities in your network to enforce constraints when finding available paths for routing between the start and end points of the required service.
Possible air-gap solutions are also suggested, if requested. An air-gap solution recommends a minor network build at a hub site to allow for a more efficient path solution. Air-gaps connect two equipment nodes in the same hub site to help form the end-to-end path. Among the configuration steps that might be recommended are:
Card fills (to be added to an existing equipment slot)
Physical straps (required to connect two pieces of equipment)
See "Routing Trails" for more information.
Network Intelligence uses a rule-based analysis engine containing an extensive list of built-in rule types to allow planners to specify policies that control how service demands are routed in the network. The analysis engine determines the optimum solution for the total service demand placed on the network based on the network model and the policies in force. You can add extensible attributes to model custom characteristics and include those attributes in determining and prioritizing routes in the network.
This planning approach is based on technology-independent services characteristic of next-generation networks. You specify high-level service demands and define how they are carried in the network; different parts of the network, based on different technologies, can carry a service in different ways. Protection rules specify the survivability of the service in the face of network outages. Over- and under-booking rules allows the setting of contention ratios or reservation of spare capacity.
See "Understanding Services, Policies, and Rules" for more information.
A service is a product offering delivered to a customer location through trails. A service typically provides a specific set of communication capabilities to a customer. for example, a Gold E-Line service at a network provider offers the fastest download speeds, highest broadband rates and lowest contention to customers.
Services are defined by service definitions, essentially blueprints for creating services, and can be grouped using a service group. A service can also reference a policy.
A policy defines a set of routing constraints, or rules, that are applied when looking for a path between two points in the network. So, for example, policies specify class of service, or quality of service constraints when looking for suitable network resources to satisfy service demand requirements on a service level agreement (SLA).
A service policy is made up of routing rules. A routing rule is a conditional constraint applied when looking for route solutions for a service demand, for example:
To limit the scope of a search to specific network elements, such as a particular network equipment provider, or particular topology;
To define requirements such as the maximum number of hops a solution completes, or the use of physical strapping to build out the network;
To generate costs associated with routing solutions.
See "Using Routing Rules" for more information.
Configuration planning supports complex network planning scenarios that require several configuration changes. It is principally used to support work order input for network engineers, such as:
Bulk trail update
Bearer or facility upgrades
Card slot movement
New service builds
New planned node builds
New transport facility or bearer creation
You capture live configuration changes to network resources in Network Intelligence (for example, site creation, equipment modification, or trail deletion) and view these actions - and interactions with other supported entities - in a configuration plan.
Configuration plans apply to the following entities:
Equipment, and the associated equipment hierarchy: cards, ports, logical ports, and so on
Trails, and the associated trail hierarchy
See "Configuring Plans" for more information.
Forecasting is the creation of optimal network build plans by network planning engineers. Network Intelligence analyzes projected service demands versus the current network capacity. The resultant forecast is defined as a collection of service demands with expected future trail growth counts for one, or more, future time periods. Each individual network point-to-point service demand consists of a route with a quantity of trails that require routing. A service demand can also have a customer. The service demands are used to automatically configure routes for multiple trails over the existing, and new planned network. The outputs of the service demand are:
Network build plan: itemizing new build requirements
Network financial budget: itemizing cost line items
Network impact: itemizing the effect of the forecast plan on network capacity
Forecasting with Network Intelligence offers several benefits:
Providing operators with the confidence to predict, and minimize, network investment requirements.
Estimating future sales forecasts by internal business units, or sales forecasts derived by Network Intelligence.
Calculating network point-to-point service demands from the plan. Individual route rules can be adjusted, and the plan incrementally rerun.
Using service demands to configure proposed routes over existing, or planned network topologies.
Calculating the amount of new network build required, and generating the cost. New network is built only when, and where, it is needed. The cost of the new build required for a sales bid can be determined. Network capacity for the bid can be reserved.
Creating a network investment blueprint to support the business in the future. Services or individual routes can be prioritized, to ensure that premium services and customers are catered for first.
Providing exhaustion analysis (trending) after implementation of planned demands. Forecasts can be continually compared to the actual take-up in the network to help improve future plan accuracy.
To create plan demands, Network Intelligence reviews past network growth to find trends that can be applied to the future. For example, by tracking how many trails are added each month, Network Intelligence can attempt to predict how many are needed in the future.
See "Forecasting Service Demands" for more information.
Network Intelligence performs impact analysis on outages and generates solutions for the restoration of affected services. Outage management is used to report affected inventory systems during a planned, or unplanned, network outage. You can also determine the outage impact on customers, sites, trails, equipment, services, cards, and other entities.
For example, if there is a planned outage for some part of your network, such as the replacement of a cable supplying customer broadband services, you can generate an outage report to find out which sites, equipment, customers, and services are disrupted, and the ways in which they are impacted. Similarly, if network services are interrupted (for example, a cable is ruptured during building works) an equivalent set of reports are generated and relayed to customers. The information in the outage reports can also help you restore the network by generating a set of service demands to suggest restoration paths for impacted services. You can also notify a list of contacts in affected customer accounts.
Outages can be tracked by date and duration for summary reporting, and this information can be saved (for example, for service level agreement tracking). You can control customer email notification by filtering the list of contacts at a client company.
See "Managing Network Outages" for more information.
Migration of network elements assists in the planning of asset relocation, premises closure, and technology upgrades.
You create a migration plan to generate network requirements changes to your network. For example, if you plan to upgrade bandwidth capacity, you can find the impact such an upgrade has on the trails in the network. Carrying out migration of network capacity facilitates the mapping of old services, sites, or nodes to new services, sites or nodes by generating a plan for network development and a decommissioning plan for entity removal, as required.
You generate a migration plan with multiple network elements of the same entity type, such as trails, and with entities of different types (for example, site and equipment). It can also be used to find the migration impact based on supported network elements.
See "Managing Network Migrations" for more information.
Network Intelligence analyses network data to identify trends in the network and predicts exhaustion dates for all network elements. For example, you can track how many trails are added to the network each month, and predict when the current equipment inventory no longer supports the required trails. You can configure Network Intelligence to email reports to selected company contacts when a set user report condition is met. For example, a notification can be sent when 60 percent of the trails available are used for a given site. The email contains a link to the report, or reports, where these conditions have been met.
Network Intelligence uses a web-based online analysis and reporting tool to provide a set of specific key performance indicator (KPI) reports. Reports of interest can be viewed in summary or at a very detailed level, and the results opened in Microsoft Excel. A network manager can define a dashboard of summary reports that quickly displays the overall network picture and provides alerts about potential network capacity problems, including network trending on all levels (for example, if service levels are in danger of being breached, this enables the network planner to requisition further capacity to cater to customer needs).
See "Monitoring the Network" for more information.
You use Network Intelligence to create routing plans for future capacity demand forecasts, typically in the short to medium term (for example, 6-18 months, against existing and already planned network capacity). The routing plan enables you to determine the minimum new capacity required, where it is needed, and how much it will cost to deploy. CSPs can narrow the gap between capacity supply and demand to enhance profitability and eliminate unnecessary capital expenditure.
Consider a CSP with current annual network growth of 15-20%.VC-12 circuits serving these capacity utilization requirements are currently modeled in Network Intelligence and added at the rate of 200-250 per week. Instead of carrying out reactive capacity management entailing significant increases in operating expenditure and the added risks of network outages, Network Intelligence capacity plans help determine possible capacity crunches within a two-year horizon at current network expansion rates.
Capacity planning replaces a wasteful network build (which eats into capital expenditure and leads to inefficiencies in network utilization), with a timely and proactive network build (that offers savings in capital expenditure and provides utilization adequate for network needs).