Inside an NBN node at Umina Beach

Poor NBN FTTN/B design may lead to decades of congestion

“With a measly 2Gbps backhaul per node you can forget about 4K Netflix. FTTN is going to be no different to the current Telstra RIMs”

(analysis) Customers on the shiny new NBN FTTN and FTTB networks may find themselves left with slow and congested speeds for decades because of short-sighted network design decisions made by the company.

Analysis of the latest Network Design Rules for the NBN, dated 30th June 2015, reveals that customers may only be able to reach a committed information rate (CIR) of roughly 5Mbps on a fully loaded node – far short of the 25 mbps that popular internet streaming service Netflix says is required for 4K video streaming and also falls short of the Vertigan panel’s recommendation that 50% of Australians will only need 15 Mbps by 2023.

The calculation:

nbn™, the company responsible for building the NBN, currently deploys 4 Point-to-Point fibres from the Fibre Access Node (similar to an “exchange”) to the NBN node where the DSLAM equipment is located.  However, it also goes on to say that only 2 of the 4 fibres will be used for connectivity, with the other 2 reserved for “future growth or migration activities”.

NBN's Network Design Document explains 4 fibres will be allocated per NBN node with only 2 in service.
NBN’s Network Design Document explains 4 fibres will be allocated per NBN node with only 2 in service.

Each of the fibres will deliver a 1Gbps ethernet connection back to the NBN Access Aggregation Switch (AAS), totaling to an effective 2Gbps ethernet connection between the node and the Fibre Access Node – or 4Gbps if all 4 allocated fibres are used.

nbn™ introduces an Access Aggregation Switch (AAS) to combine traffic from multiple nodes to the POI
nbn™ introduces an Access Aggregation Switch (AAS) to combine traffic from multiple nodes to the POI

Also according to the document, depending on the DSLAM configuration, each of nbn™’s FTTN and FTTB nodes are capable of connecting up to 384 premises.

Table of NBN Copper DSLAM options
Table of NBN Copper DSLAM options

Taking all of the above into consideration, in a worse case scenario on a  fully-loaded node at peak hour, customers may only reach 5 Mbps if all traffic was distributed evenly:

Number of DSLAM ports Fibres used for uplink Entire node’s effective uplink (Mbps) Committed Information Rate (CIR, Mbps)
48 2 2000 41.7
192 2 2000 10.4
384 2 2000 5.2

Network design wreaks havoc for binge watching season

If a mere 21% of all premises connected to a node starts streaming a 4K stream on Netflix, the node will exceed its capacity.

As many saw with the launch of popular internet TV streaming service Netflix in Australia, telecommunications companies failed to predict the demand of the service leading to heavy network congestion across Australia’s major ISPs.

For some of Netflix’s popular productions like House of Cards and Orange is the New Black, the company releases all the episodes of in their series at once.  This results in a brand new network usage “profile” that Australian ISPs and network providers like NBN have seen little of before… where customers continuously watch (binge) and subsequently stream content for hours on end.

If a mere 21% of all premises connected to a node starts streaming a 4K stream on Netflix (21% of 384 at 25 Mbps), the node will exceed its capacity.  This will leave zero bandwidth for the remaining 75% of customers potentially connected to the node.

While it can be expected that NBN’s QoS (quality of service) management will balance the load to prevent a small number of customers hogging the entire link, all customers across the board will suffer from congestion issues because of it.

With the increasing prevalence of Internet TV in Australia, the limited design of the NBN FTTN and FTTB networks will have lasting implications on what Australians will be able to do with their Internet connection.


FTTP upgrade path, uncertain

nbn™ has also indicated that they will only deploy 12 fibres up to an NBN node, making it difficult to upgrade an NBN FTTN or FTTB node area to fibre without significant downtime or extensive civil works.

nbn™ is introducing the star topology for the MTM rollout to save money
nbn™ to rollout 12-core fibre to each node, rather than the existing 36 to save money

Assuming a fully loaded 384 port NBN node is to be upgraded from FTTN to FTTP, with 4 fibres already allocated to the FTTN DSLAM for connectivity back to the Fibre Access Node, 8 fibres are remaining to potentially deliver fibre services all the way to the customer’s premises.

However, the 8 fibres will only be capable of delivering GPON services (the FTTP technology that the NBN currently uses) to a maximum of 256 premises (each fibre can be split into 32 premises, 8 × 32 = 256).

Without causing massive disruption to all customers connected to the current node, it may not be possible to transition to FTTP on high-capacity nodes other than by rolling out the network from scratch again.

This means that even if nbn™ decides to upgrade the network, they will likely continue using copper-based technologies for the years ahead to avoid large capital costs again.

(edit) further reading: You can read nbn’s side of the story in their blog post here.

CVC remains the single biggest threat to NBN

Standard 12/1 mbps plan could rise to over $150 if pricing model doesn’t drastically change

Despite the debate about technology used, the cost of NBN’s CVC remains the single biggest threat to the NBN’s success.

While the availability of high-speed, unlimited broadband grows in developed nations around the world, the future of “unlimited broadband” in Australia is becoming increasingly bleak.  The use of applications requiring greater amounts of bandwidth grow in households is partially to blame, but the country transitions to the NBN – the way nbn designs their pricing structure will also be increasingly important.  Currently, the NBN pricing structure is far too cost prohibitive for the amount of traffic the network is designed to carry.

What is this CVC thing?

Connectivity Virtual Circuit (CVC) is a virtual charge imposed by NBN to service providers to offload your traffic from the NBN network into the service provider’s network.  After a discount introduced at the start of the year, NBN charges $17.50 per Mbps of traffic shared across the ISP’s customer base.  Initially, the cost was $20 per Mbps.

This is in addition to the cost of physical interconnect connection between the provider and NBN (called the Network-Network Interface, NNI) plus the cost that NBN charges for the link between your home and NBN’s point of interconnect (known as the Access Virtual Circuit and User-Network Interface, AVC/UNI).

The Netflix effect

It became pretty obvious after the launch of Netflix in Australia that the current pricing structure is unsustainable.  With the high-bandwidth of the NBN, customers expect to stream movies and TV shows with plenty of remaining capacity to do additional work in the background.

To be able to deliver a HD stream, Netflix recommends 5Mbps of bandwidth.  To be able to guarantee at least a single stream to every household, ISPs need to purchase 5Mbps of CVC per household:

$17.50 × 5 Mbps = $87.50

This neglects costs like the actual cost of connecting you to the Internet (Backhaul/IX costs) and link between your home and NBN’s point of interconnect – which start at $24 for 12/1 Mbps. Backhaul and IX costs depend on the ISP’s deals with backhaul providers and the volume they have – but consider $10 per mbps a reasonably conservative estimate. So, for a typical 12/1 connection that can consistently deliver at least 5 Mbps:

Cost element Cost
CVC – 5 Mbps $87.50
AVC/UNI – 12/1 Mbps $24.00
Backhaul/IX Costs – 5 Mbps $50.00
Total $161.50

Since a typical 12/1 NBN plan costs far less than the $161.50, you can understand where the compromise lies –  CVC, IX and backhaul.

While you wouldn’t necessarily expect that all users will simultaneously watch Netflix at the same time, the number of simultaneous streamers is growing exponentially.  ISPs also need to prepare for “extraordinary” events like when Netflix released the entire season of Orange is the New Black – a far greater proportion of users will be watching Netflix at the same time for an extended period of time.

Future of small providers

With the high CVC costs, small providers will find it increasingly hard to compete with larger providers like Telstra, Optus, iiNet and TPG.  These providers either have enough traffic volume (or own their own backhaul infrastructure) to lower costs in the backhaul or IX component to offset the high CVC.

Smaller providers still need to “rent” capacity from backhaul infrastructure providers such as Telstra, AAPT, PIPE and Nextgen Networks.  With even large providers like iiNet feeling the pinch, it’s no wonder that smaller ISPs are beginning to struggle.  Even with the scale of a large customer base, current consumer cost expectations are simply unsustainable.

What’s a solution?

nbn™ could continue dropping the CVC charge as demand increases, however, it faces the risk that service providers will continue to skimp out on CVC – thus lowering revenue.

The good news is the company has also begin conducting a second round of industry consultation to help overcome the problem.

For me, one solution I can see feasible, is to replace the CVC cost with a standard capacity charge (which guarantees a certain ratio of CVC to AVC) comparable to current industry contribution to CVC per user.  Then, provide capacity boosts for users and applications that require a greater capacity.  This guarantees “minimum” revenue for nbn™ and increases capacity to suitable levels for service providers.

Alternatively, I could see increasing the user-connect costs (AVC/UNI) and drastically cutting CVC costs as a possible solution.  This would also guarantee revenue to a certain extent, while retaining the control of contention with service providers.

Whatever happens, the nbn™ pricing model needs to change drastically in order for Australia to remain competitive in the 21st century.  Otherwise, we’ll continue to be slowed down by a “virtual” cost-prohibitive charge by our nationwide broadband network.

Netflix launches in Australia with 1k titles!

Netflix has finally launched its Australian and New Zealand offerings. Needless to say, one of the first questions was “what movies/TV shows are being offered”?

Based on my initial number crunching, it looks like a total of 1,326 titles are being offered by Netflix in Australia. I’ve compiled a spreadsheet that lists all titles I was able to get from the Netflix site. There is also a US vs CA vs AU comparison on the second sheet of that spreadsheet.

Here’s a quick comparison grid for Netflix in Australia, the US and Canada.

Common titles in AU v US v CA
AU 676* ~145 ~138
US ~145 ~5,649* ~2,291
CA ~138 ~2,291 ~1,567*
AU+US+CA ~367 (common in all 3)
Totals in each countries
Total 1,326 ~8,500^ ~4,000^

* These are unique titles — titles are are only available on that country, and not the other two in this comparison
^US and CA titles are obtained from Netflix Canada vs USA list

Updated at 10:27AM with revised estimates: AU list is now broken down into seasons — this should fix things up a bit. Note, figures are not hugely different. Old figures are at the bottom of the post for comparison.

Please still take these figures with a grain of salt, and take note of the disclaimer below.

Update at 9:00AM: It has come to my attention that the comparison between AU and US/CA isn’t as easy as matching the title IDs as I have done… because of the way the data was extracted. The AU list was obtained by getting a list of titles (for TV shows, it’s the series name) – where as the US/CA list is broken down into each season of the TV show.

An updated spreadsheet is coming soon. Apologies for the inconvenience!

As usual, it looks like Australians are still getting less titles available. One can only hope the list of titles offered by Netflix in Australia will increase as time goes on.

Disclaimer: the numbers are estimates and for reference only. There is no guarantee of their absolute accuracy and may only be good for “order of magnitude” approximations.