Network Planning Ready for a Starring Role

In the classic movie, The Graduate (1967), the protagonist is advised on career choices, “In one word - plastics.” If you were asked by a young person today, graduating with an engineering or similar degree about a career choice in telecommunications, would you think of responding, “network planning”? Well, probably not.

Why? Because… let’s face it. Planning has always been considered boring - one of those tasks that took place in the background at a leisurely pace. You started with existing traffic, created forecasts for growth, created a few network model variants based on current architectural approaches using a decade-old planning tool, and then produced management reports for capital budgeting purposes. Slow, predictable, little room for innovation. If you were starting a telecommunications career, it would be much better to focus on cutting-edge products that handle traffic faster and more efficiently.

This situation is beginning to change, for several reasons:

• There is much more at stake. Traffic is continuing to grow at a breakneck pace. Planning errors could be more costly than ever.
• Services are becoming more varied, making networks more complex. Particularly with the onset of 5G, the network must be planned to support multiple types of services with different performance characteristics.
• Service assurance is no longer a synonym for issuing rebates when there is an outage. True assurance means ensuring sufficient network redundancy or restoration mechanisms to deal predictably with failures.
• Economic pressure is greater than ever. Network operators no longer have the luxury of dealing with uncertainty by throwing additional resources at the network. They need to get the most traffic handling capacity for the minimum investment. This same pressure is finally forcing them to take a harder look at planning optimized multilayer networks.

As a result, there is a definite shift in emphasis in telecommunications, from the products to the network. And even more, in being able to plan networks that meet all of the listed concerns, with confidence. This requires a fundamentally new approach to planning and planning tools, as follows:

• Work on actual network configurations: For brownfield upgrades, the planning tool should not have to rely on records. This requires mechanisms that extract actual status directly from the network.
• Develop better algorithms: Network planning is hard to do, mathematically speaking. It falls into a class of problems, like the traveling salesman problem, called NP-hard. These are really difficult. This means that as the network increases in complexity (e.g. more nodes) it can take an exponentially longer time to approach an optimal solution. There is huge scope for modeling innovation.
• Incorporate multilayer considerations: At a basic level, the planning tool must ensure that logical route redundancy at the service layer is not nullified by physical implementation of these routes running for a portion in a common fiber bundle. At a more advanced level, multilayer optimization should aim at reducing the number of expensive router ports through bypass architectures that switch traffic at the more economical optical layer.
• Plan for failures: The planning tool should incorporate “what if” simulation capabilities to analyze the impact of a node, link, or even a card failing. This allows preemptively addressing potential problems, particularly if the impact is not meeting guaranteed SLAs.
• Continuous optimization: To keep pace with a faster world, network planning is moving from cyclic upgrades to continuous optimization. Triggered by analytics that flag ‘traffic pain points’ in the network, planning tools should direct continuous reshuffling of existing resources to optimize the network’s traffic-carrying and automated failure recovery capabilities. Moreover, when no amount of reshuffling helps, the planning tools should initiate ordering new resources to expand the overall capacity.
• Plug-and-play network upgrades: In the same way that planning extracts actual network configurations, it should be able to work hand-in-hand with network commissioning to ensure fast and error-free installation of new network equipment.

Well, I think you get the idea. Network planning is becoming increasingly important, and even exciting. It is ready to take on a starring role and shine. In the spirit of Flashdance (1983) it will “take your passion, and make it happen.”

At ECI, we are taking these messages to heart with Muse^TM Network Planner. Whether for greenfield or brownfield deployments, it works on real network data, creates plans and continuously optimizes, uses cutting-edge optimization algorithms, and features “what-if” simulation tools to predict how the network will react to unforeseen events.