Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Bowdon]

The main complaint about the current situation is that OR are favouring BT's customers, and not giving proper information or responses to other ISP's. It would be interesting to see the evidence for this. If the evidence is there then it needs to be asked why this is happening, and how it can be improved.

I do think BT are the more qualified in this technology market and so would like to see BT remain at the helm of OR, though I would like to see some kind of deal done so that ISP's can be allowed more access to OR (maybe access to DLM control like what is currently done for ADSL). Openreach should be made to answer questions when asked and not have this 1940's mentality that everything needs to be a secret.

If the secrecy lid is lifted and other ISP's are allowed more direct control over their customers lines, and also be able to get answers from OR, then I would also like to see more ISP's investing in Openreach, which would help fund, and maybe quicken the release of new technologies.

Imagine if Openreach had access to more funds from other ISP's. Surely that would benefit everyone?

Thats my 2 penny's worth.

[Black Sheep]

What I personally believe is that the council funds should have gone into companies that weren't BT.

This would not only create greater competition, it would actually accelerate BT's FTTC roll-out.

Read up on it, Alec. You'll see that anybody and everybody was invited to tender their bids for council-funded FTTC work (BDUK). You'll also see that there was ONLY BTOR left at the very end, especially when the predictive models were showing an average of 20-25yrs before the investment would start to show a profit !!

This however, is not even related to the potential splitting off of BTOR ............. it's a different topic. 

[loonylion]

What I personally believe is that the council funds should have gone into companies that weren't BT.

This would not only create greater competition, it would actually accelerate BT's FTTC roll-out.

Read up on it, Alec. You'll see that anybody and everybody was invited to tender their bids for council-funded FTTC work (BDUK). You'll also see that there was ONLY BTOR left at the very end, especially when the predictive models were showing an average of 20-25yrs before the investment would start to show a profit !!

Fujitsu pulled out stating that the bidding process was unfairly biased towards BT.

[Black Sheep]

In what way ?? Sounds more like sour grapes to me. It's the easiest thing to do to apportion blame with no substance ............. what about the other bidders, why did they pull out ??

[GigabitEthernet]

From what I recall, Fujitsu wanted to roll-out true FTTP to something like 75% of the UK but they wanted to use BT's ducts and BT wouldn't let them.

They also couldn't roll it out via telegraph poles as BT owns those too.

There was something else too about more meetings being held with BT and that essentially their plans were preferred. Fujitsu felt cheated so they pulled out.

[Black Sheep]

If that is the case, it's hardly surprising now that BT wouldn't let them use their ducts and poles.

I haven't the time or energy to go looking back through the posts within here, about this particular topic. It's been covered in all its gory detail many times. This thread is more to do with hiving off OR from BT Group, and as has been pointed out earlier ….. we'd be lucky to even have FTTC if that had happened years back !!.

[kitz]

They definitely were allowed access to the ducts and poles -  BT even reduced their prices to give Fujitsu & other Telco's access

The cost for using a drop pole – "the most common final pole that can connect multiple premises" – has halved from £21 per attachment to £11.

Analyst firm Ovum has studied pole and duct pricing schemes across Europe, and says that OpenReach's new pricing structure "is up to 21 per cent below the European average in urban areas."
"Crucially, in the more rural areas, [the new pricing] is as much as 38 per cent below average when considering pricing in France, Portugal and Spain," it says.

Earlier this year, IT services provider Fujitsu announced plans to install new fibre-optic cables connecting rural areas of Britain. It promised that this would allow the likes of Virgin Media and Talk Talk to offer 'superfast' broadband in communities not served by BT.

However, the small print of the announcement revealed that Fujitsu's plan would only go ahead if BT lowered its pole and duct prices.

A spokesperson for Fujitsu told Information Age today that the company has yet to make a decision on whether the new pricing makes the fibre-optic cable plan viable, but that it is "examining BT's new prices very carefully"

- See more at: http://www.information-age.com/technology/mobile-and-networking/1660383/fujitsu-%22examining%22-bts-new-pole-and-duct-pricing#sthash.39X8nvSk.dpuf

I know Fujitsu didn't even bother to bid for some of the early areas such as Cumbria because they said it wasn't profitable.  They didn't bid in a few of the early areas purely down to cost return, even before they pulled out completely.    Ive no idea if they did feel cheated or not, nor if there was some bias going on.

IMHO the whole BDUK process was flawed and too much money wasted in the wrong areas of bureaucracy.   Councils were left to do their own research and were either led by someone who knew little about the subject or technology or spending lots of money on hiring 'professional consultants'.    So you have a band of people in local council who dont have a clue about broadband who make the decisions.   

[WWWombat]

Like Kitz, I think there are swings and roundabouts here.

If we look at Openreach in terms of the old stuff - the copper connection for a voice service - then the recent graphs from Ofcom (via TBB) show that this is a dying business:


In that arena, perhaps Ofcom need to be reducing regulation and overheads. Not adding anything. And certainly not causing BT to spend huge amounts to separate back-end control systems etc.

For me, the whole emphasis on the approach to take with Openreach should be all about the future. Not necessarily to force FTTH deployment, but to set the best environment for investment; all to allow data speeds to be what people need.

Only once they've established an environment that allows good investment for 90% of the country do they gain the mandate to add regulation to force consideration of the remainder.

Right now, we have BT Group's actions (on FTTC) and promises (on FTTdp) to show the ongoing investment possibility for Openreach. If we're going to split Openreach off, we need to see a much clearer picture of how the likes of Sky and TalkTalk, and rBT, can trigger enough investment into Openreach. Should Ofcom be enforcing ISP's to invest (at least larger ones) much like Vodafone suggest that they would do? If so, who should own that Openreach?

Without enforcement, my worry is that the desire for Sky and TalkTalk is to make things cheaper at the expense of better (Sorry, Sky, I don't believe). However, the government, the future economy, really needs things to be better, not cheaper.

Enforcement sounds like the creation of an enforced monopoly, much like Australia's NBN. What would VM say about that situation?

And, having mentioned VM, I ponder this too: If Openreach gets split off because they are a monopoly (though technically to only one-third of the country, the expensive third, post-project-lightening), should VM be left as the only vertically-integrated operation, with an automatic advantage in the more competitive two-thirds?

We need to ponder carefully whether the one bird in hand is better than two in the bush. It'd be an expensive way to discover the one bird you let go happened to be the goose laying the golden eggs...

[GigabitEthernet]

There is one scenario in which Openreach is like a co-operative in that all the ISPs invest into it and in return they get use of the infrastructure.

The investment then goes to maintaining and upgrading the infrastructure.

[kitz]

Without enforcement, my worry is that the desire for Sky and TalkTalk is to make things cheaper at the expense of better (Sorry, Sky, I don't believe). However, the government, the future economy, really needs things to be better, not cheaper.

I'm also dubious about the fact that the major player requesting splitting off of Openreach is Sky.  The Murdoch empire certainly isnt whiter than white and has an interesting history.   Im also mindful on how they managed to push out any valid competition when it came to retail satellite.  Oh it was great when there were price wars, but look what happened when they won and pricing then went through the roof (literally!).  iirc things went from £5.99pm to £15.99pm and now also having to pay extra for things such as sporting events which were previously included in the package, in just a few short years.  It was at this point I ditched Sky because all I mostly watched were the Discovery type channels.

My fear is that similar will eventually happen with broadband.  Already there is very little competition these days outside of the main four.   If you want quality service then you do appear to have to go outside of those 4.   Yet 10 years ago there used to be dozens you could chose from.

[Chrysalis]

it already happens with broadband via line rental sadly.

So both sky and BT play the same game now.

Do sky do things on the cheap? there is defenite evidence of that mainly with equipment they supply, cheap set top boxes and the sky hub is a low end piece of kit.

On the other hand sky have more uk based support staff than BT and a much better fiber network, sky's LLU network as well as its peering is very good, perhaps the best in the country.

[WWWombat]

The question of how good Sky's core network is, compared to BT's is a very good one ... and probably not directly answerable by mere members of the public.

However, probably by chance, there were a couple of presentations at UKNOF earlier in the year that gave a hint of a comparison. One was a presentation by Tim Rossiter of Sky, and the other by Neil McRae of BT.

Sky's presentation described their newly-upgraded core network (with 4 core locations; two in London, one each in Birmingham and Leeds), including the migration process and the progression to 100Gb optics.
https://www.youtube.com/watch?v=uQUoO4Wb7s4

It sounds like a fairly meaty network, both router- and optical-wise, but it looks like the three (paired) routes down to London become an absolutely critical leg (in the questions, Tim mentions that most traffic comes into the south, and I suspect an awful lot of their total traffic gets squeezed down these legs).

Total traffic was running at 3Tbps at the start of 2015 (possibly up to 5Tbps by the end of 2015, 65-70% being online video), and the new core nodes (paired at each core site) are each capable of 8Tb, with a roadmap of up to 16Tb. Because of the recent upgrades, the core network doesn't sound overloaded to me, but there are perhaps possibilities every few years as new migrations will be needed. Sky are running on Cisco's bleeding edge, technology-wise, and in my (entirely amateur) opinion could do with some more diverse routes down the country and/or some more connections to the outside world in the north.

With the 4 core sites, there are 65 PoPs, each feeding (I guess) around 50Gbps into that core network, mostly via add/drop multiplexors on the inter-core routes.

However, I believe Sky need a good core network for a reason - that all being about their cash-cow of TV content. While I believe they see "broadband" as the cheap loss-leader to compete with VM & BT, they want their TV content to bring in the money. And I suspect they'll want to build up their IPTV capability more and more to work as an alternative to satellite distribution.

So ... I'm willing to believe that Sky are OK with investing in decent core network equipment, but less willing to believe they'll invest in the access network as much. If they did that, they'd be helping the competition too much.

BT's presentation doesn't tell you much about the capability of the core network, as it focusses on the new fibre build for the Scottish Highlands & Islands.
https://www.youtube.com/watch?v=5ZU5tvBUwZM

However, it does give you a feel for the capabilities right at the edge of the core network: They mention having 10x10G, 100Gb, lit around the new parts of the fibre network. It appears this is in 4 separate legs (so I guess 10x10G along each leg), and the new legs seem to sprout from existing metro nodes around Inverness, Dundee, Glasgow and (somewhere near) Kilmarnock. It shows 30 new WDM nodes spread along these legs, which (if Kitz' 21CN article still applies) probably promotes 30 medium exchanges into being Tier 1 MSANs.

From the 21CN architecture, we know BT have just over 100 Metro nodes; these 100 nodes will have multiple legs (of 10x10G, or presumably more for the more urban routes) that feed into the 20 core nodes, so I guess the overall capability could be huge. During Q+A at the end of the UKNOF presentation, Neil mentions that every Metro node has a single hop 100Gbps connection to London ... suggesting that metro <-> core capacity runs at over 10Tbps.

The original Kitz page on 21CN shows 10G links from Metro nodes to the Core (and within the meshed Core too), but I guess things have definitely moved on on this front. I also wouldn't be surprised if the overall linking architecture has changed somewhat; in the original days, the fully-meshed core would have been important for the VoIP component. Nowadays, the internet component will dominate - so short-cuts to the inner-core internet PoPs would make sense.

For total utilisation, a different presentation (mere PDF) suggests that BT are predicting 12Tbps traffic in the core network by 2020, and are talking of compound growth rates of 45% leading to that ... which suggests it is currently around 2 Tbps.

I pondered why Sky would have a higher amount of core network traffic compared to BT, and a higher growth rate, and the best answer I can come up with is that a good chunk of BT's customers will be using multicast TV streams. However, a good proportion of Sky customers will be using NowTV streams that will be unicast.

For comparison, TT hit a 1.4Tbps peak around the end of last year.

[Chrysalis]

yeah I more meant how hot both companies run their networks, it wouldnt surprise me if BT's network was larger, but with more points of visible contention.  So basically by better quality i mean sky have better capacity planning.

Both networks appear to have redundancy as I noticed with sky when I auth, it affects base latency for the duration of the session, the difference isnt big but its visible, I will either get 8-9ms or 9-10ms to london depending on the route.  BT wholesale we know has redundancy by the same measurements but also by the fact one can avoid the congeston by getting alternate routing.

For me sky also has the additional benefit that they have a POP in my city (hence my low latency) meaning my traffic goes directly from here to london.

i expect sky has basically ex easynet staff running their network, and easynet had good policies and a good network when brought out by sky.

Also ignition posted elsewhere how good in terms of capacity sky's public peering is when compared to number of customers.  Although in this case I dont want to include peering into the measurements just the core network.

[kitz]

The original Kitz page on 21CN shows 10G links from Metro nodes to the Core (and within the meshed Core too), but I guess things have definitely moved on on this front.

They did a massive upgrade and overhaul of the 21CN core last year.
They heavily invested many hundreds of Alcatel 7750s -  not just for the MSE's either.  These are capable of taking 400Gbps of traffic on to the Core.

[kitz]

Also ignition posted elsewhere how good in terms of capacity sky's public peering

Its their peering that they are good at.   Its something rizla and I have discussed in the past.  They inherited a lot of good transit and peering when they took over UKO.