NBN network design dissected

In The Australian today the focus is on how the company building the National Broadband Network will be hard-pressed to defend aspects of its network design after the most recent update to the Network Design Rules (NDR). The decision by the government to force NBN to use a lowest cost rollout approach has meant that company finds itself adopting technologies and a network design that will be outdated even before the rollout is completed.

Read the full article below

The company building the National Broadband Network will be hard-pressed to defend aspects of its network design after the most recent update to the Network Design Rules (NDR). The decision by the government to force NBN to utilise a lowest cost rollout approach has meant that company finds itself adopting technologies and a network design that will be outdated even before the rollout is completed.

The “Fast. Affordable. Sooner.” mantra professed by the Coalition is underpinned by the principle that the federal government should not have taken a direct role in the provision of broadband. Consequently, the current approach is likely to provide little more than anincremental upgrade to the existing broadband access networks.

Key technology choices and how to deal with reliability, resiliency and capacity are now in the spotlight and while the NDR documents provide a fairly lightweight technical description of the network design there’s more than enough ammunition here for proponents of the “do it once, do it right” Fibre to the Premises (FTTP) approach.

NBN's chief architect Tony Cross describes in a blog the challenge that the company faces in building a network using the “most cost efficient manner possible and making the best possible use of taxpayer dollars.”

What Cross fails to provide is justification for the statement and what NBN is doing other than to minimise upfront rollout costs as required by the government’s latest Statement of Expectation. NBN has not carried out a life cycle cost and performance analysis for the anticipated 30 to 50 year lifetime of the network. 

Cross argues that “the reality is that we can quite easily upgrade our FTTN cabinet backhaul capacity whenever we need to do so by simply installing a new optical interface – there is no need whatsoever to run any new fibre.”

But what does “quite easily upgrade our (Fibre to the Node ) FTTN  cabinet backhaul capacity” actually mean, how much is it going to cost and will this cost be included in the Total Cost of Ownership (TCO) projections for the period to when the NBN rollout is completed? Is there a risk that to ensure the TCO is minimised upgrades are delayed until the network is disaggregated?

Brevity and missing information

The brevity of the NDR makes it difficult to clearly see many aspects of the technical design surrounding the FTTN, Fibre to the Basement (FTTB) and HFC networks and this is arguably because there are aspects of the introduced technologies that are only now becoming available, and this particularly applies to HFC and point to point fibre connections.

From what is provided it's difficult to gain a clear understanding of the HFC network design including how many premises will be connected to a cable run, how much of the current fibre in the HFC network will be leased from Telstra and Optus, and what NDR modifications will be necessary to accommodate DOCSIS 3.1, which has been announced by the Communications Minister Malcolm Turnbull and NBN for implementation commencing in 2017. Similar concerns exist for the point to point fibre connections.

Cross goes on to state that “The other complicating factor that we have is that in some parts of the network we currently have to lease backhaul capacity from private sector operators.

When we do this we have to pay on the basis of a peak bitrate required – this means we have to be very prudent in working out exactly how much demand will be required or face the prospect of paying a lot of money for capacity that we did not actually need.

To make sure we do this properly we continuously monitor our traffic levels and we increase our capacity when we need to do so to help us maintain our quality of service – but we make that investment only when it is required.”

Cross identifies the need to minimise leased backhaul capacity costs but this requirement is distinct from NBN’s decision to install access network equipment with a traffic capacity capability that is unlikely to meet customer demand for very long and the cost of replacement will substantially reduce the hotly debated rollout cost differential between FTTN and FTTP.

Upgrade costs still obscured

The NDR document provides a strong justification for NBN’s critics to demand the publication of the anticipated costs that would be incurred to reduce congestion and provide improved performance including upgrading the DFN to two 10 Gbps Ethernet connections, the cost of the announced 2017 DOCSIS 3.1 upgrade to HFC and the anticipated soon to be announced pre-2020 upgrade for FTTN from VDSL2 to G.Fast.

There appears to be a rollout cost differential between the MTM methodology and the FTTP methodology of about $500 -$800 per premises excluding the premises drop line and internal fittings (about $500) which could arguably be pushed onto customers through retail service provider product plans. NBN needs to justify why a national access network replacement project costing $43 billion should have a lifetime of about two to five years before upgrades are necessary and before the rollout is completed.

The projected $1 billion per annum additional OPEX cost associated with the legacy networks included in the MTM NBN further reduces the viability of the decision to rollout FTTN.

A key concern for taxpayers should be that the network is built  without it becoming a constant drain on the government’s purse and that means it would be prudent to build the network now with a 30 to 50 year lifetime.

Consumption and demand

Cross describes NBN's team as having "many years of experience in designing and building broadband networks – so we understand consumption is increasing and we know how to plan for and deal with that." 

"With this in mind it’s also very important to understand exactly how that consumption is going to occur – not everybody is going to be streaming Netflix video content at exactly the same time, the demand will be spread over the day – which is crucial to understand when planning capacity," he adds.

"In addition, it’s worth remembering that when it comes to streaming video like Netflix that end-users are not all synchronised in their viewing – this means that they are not going to be hitting the backhaul link simultaneously so you are unlikely to have capacity issues even at peak usage times.”

The focus on Netflix and entertainment products underscores concern engineers and scientists have about the government’s decision to provide a “bare-minimum broadband” NBN at a time when there will be significant growth in online health, education, social media and other products and applications and demand for improved reliability, quality and performance.

We need to move beyond the narrow thinking about entertainment products and to realise that in the future everything will be connected to the network and consumers will expect congestion, poor quality and poor performance to be a historical fascination.

While the revised NDR provides some answers to what we should expect by 2020 the brevity of the NDR and the missing information leaves many questions unanswered and we should expect yet another revised NDR next year.

Mark Gregory is a senior lecturer in the School of Electrical and Computer Engineering at RMIT University.