Migration to an optimized Business Driven Network (Part 3)

Posted on: 8th December 2015

This final blog, in a series of three, discussing migration to an optimized, business driven network looks at the revenue opportunities that can be gained from making phased improvements to a network.

Revenues, Business Models and Optimization

Calculating revenue in the network is complicated, as the revenue per bit being transported across a network, can vary by a significant number of orders of magnitude.


In parallel, designing a network for optimizing one set of information, in line with another set of data, will be problematic if you are attempting to do so without the knowledge of exactly what the required services are, and what they need to deliver on based on business objectives and the revenues associated with them.

Business models have changed and will continue to do so, adding more complexity.  Service providers would prefer to price ‘per megabyte’ of information but the growth of flat rate data plans realistically rules that out.

Designing a network around a business model is critical.

If operators want to achieve business driven optimization – where the shape and configuration of the network is directly driven by overall business goals – then network planning and design becomes central to balancing revenue, OPEX and CAPEX in order to optimize margins.

The Evolution of Optimization

Service providers have always wanted to be efficient in their use of resources, and their approach has developed over time. Business-drive optimization represents a significant evolution relative to the approaches taken to date:

Asset driven optimization – which involves looking at the equipment in the network, trending it, and predicting future need – is a very different concept. It is a network-centric view of planning and optimizing a network, and whilst valuable it has little to do with the revenue generating element of the business.

Service driven optimization is somewhat more useful as it looks at how a service is overlaid on a network, and the resources that a service needs in its deployment across a network. This is closer to the money because it carries revenue but does not meet the requirement of meshing business objectives with network planning.

Let’s see how these approaches compare.

Optimization Examples Compared

Consider a competitive access network provider in a metropolitan area interconnecting with a national PTT.

If the network provider was looking at managing its network through asset-driven optimization it may opt to trend port counts over a certain period of time and acquire the correct port density on a two quarterly planning horizon.  The service provider would then know if it needs 3G’s worth of capacity, in two quarters time, rather than deploying 1G.

A service-driven optimization approach may instead trend the service demands to see how these services are deployed across the network, alongside capital and operational costs in order to understand what those services would require to be delivered on the network and to get the best value without violating SLA’s.

Business Driven Optimization goes one step further.  The objective here is to optimize equipment going into the network, or the OPEX, but also closely monitor the service revenue – namely, the interconnect agreements and cost differences between leased capacity and bespoke builds.

The biggest difference from the operator’s perspective is that this is a fundamental shift from being reactive to being proactive and enables a service provider to start looking ahead to being margin focused. It moves from being very parochial, looking at the network appliance itself, to being service focused and driving self optimization.

A further level of complexity is provided by constraint based routing, deciding how services path across the network, automating at multiple layers and automatically feeding and optimizing. This is where SDN comes into play.

By looking at the network in terms of service change and bringing services together, automatically rerouting and recreating functions in the right places, it is possible to appreciate the true business value of an AI-enabled network and why operating in line with business objectives, rather than in spite of them, is a valuable end game.

Operating a network that knows exactly what would happen and where, under different failure scenarios, is a useful route to driving up revenue opportunities.

Whenever Aria Networks talks to a customer, or an operator, it is ultimately clear that the CEO or CFO wants more control over margin and network management.  SDN enables this to become a reality.

Read part 1 and part 2 in this series