User comments on ISPs
  >> Virgin Media (Cable)


Register (or login) on our website and you will not see this ad.


Pages in this thread: 1 | 2 | 3 | 4 | 5 | [6] | 7 | (show all)   Print Thread
Standard User Ignitionnet
(knowledge is power) Thu 12-Apr-18 13:32:27
Print Post

Re: Thinkbroadband Speedtest capped download speed?


[re: cje85] [link to this post]
 
In reply to a post by cje85:
According to PeeringDB VM now have 2x 1TB links at LINX. Liberty Global have an additional 200Gb.

https://www.peeringdb.com/net/1412
https://www.peeringdb.com/net/346


According to LINX VM have 6 x 100G split 4-2 to the LINX public LANs. Liberty have a single 100G into each.
Standard User kitcat
(experienced) Thu 12-Apr-18 16:45:25
Print Post

Re: Thinkbroadband Speedtest capped download speed?


[re: Ignitionnet] [link to this post]
 
cje85 / Ignitionnet

I am sure you know ( or at least 1 of you does) that physically VM have 2 TB links but logically/optically are only using 6 100Gb of that capacity, split 4 & 2. So you are both right!

So is jchaimer who gives the reason as cost. Which is why they have not equipped the other capacity. They should be able to use 10 100Gb pipes/LANs on each 1TB physical.

As each TB is likely to be an optical system, only equipping some wavelengths saves a significant amount of cost Not just in optics but also in router ports and cooling.

It is highly likely that they will not be able to use all 10 100Gb per TB due to the cooling overhead on the rack. When the high density Optics first came out quite often only 1 wavelength was lit and the specs, if you equipped them all, described the heat output of a large bonfire not electronic equipment ( 20Kw per box and you could fit 3 boxes in a rack if you were careless enough!)
Standard User Ignitionnet
(knowledge is power) Thu 12-Apr-18 17:09:24
Print Post

Re: Thinkbroadband Speedtest capped download speed?


[re: kitcat] [link to this post]
 
In reply to a post by kitcat:
cje85 / Ignitionnet

I am sure you know ( or at least 1 of you does) that physically VM have 2 TB links but logically/optically are only using 6 100Gb of that capacity, split 4 & 2. So you are both right!

So is jchaimer who gives the reason as cost. Which is why they have not equipped the other capacity. They should be able to use 10 100Gb pipes/LANs on each 1TB physical.

As each TB is likely to be an optical system, only equipping some wavelengths saves a significant amount of cost Not just in optics but also in router ports and cooling.

It is highly likely that they will not be able to use all 10 100Gb per TB due to the cooling overhead on the rack. When the high density Optics first came out quite often only 1 wavelength was lit and the specs, if you equipped them all, described the heat output of a large bonfire not electronic equipment ( 20Kw per box and you could fit 3 boxes in a rack if you were careless enough!)


I'm probably misunderstanding this quite grievously.

Into the LINX public LAN VM physically have 6 x 100Gb router/switch ports connected to 6 x 100Gb Ethernet switch ports across 6 different switches on the two LINX LANs.

Logically and optically VM have 6 x 100Gb Ethernet router/switch ports connected to 6 x 100Gb Ethernet switch ports on 6 different switches across the two LINX LANs.

Connections to the LINX LAN are over 1/10/100Gb Ethernet only. Nothing to do with WDM at all, beyond the wavelengths necessary to carry 100Gb Ethernet.

PeeringDB just have this wrong. It happens.


Register (or login) on our website and you will not see this ad.

Standard User kitcat
(experienced) Fri 13-Apr-18 13:15:08
Print Post

Re: Thinkbroadband Speedtest capped download speed?


[re: Ignitionnet] [link to this post]
 
I was just pointing out that due to partial truths both are correct, however the PerringDB info is misleading.

PeeringDB have most likely picked up on the 2 TB ( Optical systems) into the building which VM then split up before connecting at 100Gb to the switches and onward..

The 2 TB systems will ( should ) be diverse which is why they have 2.

It is quite normal for Operators to 'claim' the capacity as the optical capacity as this looks good, but only light less capacity as the switch / router ports cost.

I can remember some claiming 64 fibre pair connectivity but only using 1 pair. Made the connectivity seem really good.
Standard User Chrysalis
(legend) Sat 14-Apr-18 00:45:03
Print Post

Re: Thinkbroadband Speedtest capped download speed?


[re: cje85] [link to this post]
 
yes, in case it wasnt obvious, I wasnt saying they have never upgraded since then. smile

Sky Fibre Pro BQM - IPv4 BQM - IPv6
Standard User Ignitionnet
(knowledge is power) Tue 17-Apr-18 13:37:05
Print Post

Re: Thinkbroadband Speedtest capped download speed?


[re: kitcat] [link to this post]
 
In reply to a post by kitcat:
I was just pointing out that due to partial truths both are correct, however the PerringDB info is misleading.

PeeringDB have most likely picked up on the 2 TB ( Optical systems) into the building which VM then split up before connecting at 100Gb to the switches and onward..


Apart from that they specifically claim it's the public peering capacity, they have absolutely no way of knowing how much capacity VM have into the routers VM have with LINX facing interfaces, and there are more than 2 VM routes into the LINX sites, sure.
Standard User nemeth782
(committed) Tue 17-Apr-18 13:54:00
Print Post

Re: Thinkbroadband Speedtest capped download speed?


[re: jchamier] [link to this post]
 
In reply to a post by jchamier:
Lots of my colleagues would pick a 300 megabit home service with variable latency and lack of single threat throughput instead of a 50/60 megabit service that has constant latency and achieves fast downloads in one thread.

General public versus those whom know a tiny bit about networks.


But that's not the whole story.

I can choose between a 350mbit home service with a stable ~22ms latency and a single threaded throughput of ~80mbit, or an FTTC home service with yes, less latency but never more than 40mbit regardless of number of threads.

Given that the single threaded speed of the 350mbit connection is still double the entire capacity of the FTTC line, it's still a nobrainer to me.
Administrator MrSaffron
(staff) Tue 17-Apr-18 14:08:02
Print Post

Re: Thinkbroadband Speedtest capped download speed?


[re: Ignitionnet] [link to this post]
 
1  thinkbroadband-gw2.core-rs2.thdo.ncuk.net (80.249.99.113)  1.414 ms  1.605 ms  1.800 ms
 2  po4-31.bdr-rt3.thdo.ncuk.net (80.249.97.86)  0.526 ms  0.513 ms  0.496 ms
 3  te2-1-9.star10g.bdr-rt1.thn.ncuk.net (80.249.97.18)  0.368 ms  0.371 ms  0.383 ms
 4  linx-gw1.router.ntli.net (195.66.224.22)  1.369 ms  1.362 ms  1.343 ms
 5  * * *
 6  * * *
 7  * * *
 8  leed-core-2a-ae4-0.network.virginmedia.net (82.15.94.254)  8.084 ms  8.046 ms  8.030 ms


 1  thinkbroadband-gw2.core-rs2.thdo.ncuk.net (80.249.99.113)  0.659 ms  0.867 ms  1.081 ms
 2  po4-31.bdr-rt3.thdo.ncuk.net (80.249.97.86)  0.549 ms  0.538 ms  0.521 ms
 3  te2-1-9.star10g.bdr-rt1.thn.ncuk.net (80.249.97.18)  0.344 ms  0.356 ms  0.388 ms
 4  linx-gw1.router.ntli.net (195.66.224.22)  1.386 ms  1.393 ms  1.373 ms
 5  * * *
 6  lutn-core-2a-ae3-0.network.virginmedia.net (62.254.42.141)  2.644 ms  2.723 ms  2.709 ms


Two traceroute just done to somewhere in Leeds and another in Luton, so will be interesting to see if things change during peak time period

The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
Standard User Ignitionnet
(knowledge is power) Tue 17-Apr-18 15:07:10
Print Post

Re: Thinkbroadband Speedtest capped download speed?


[re: MrSaffron] [link to this post]
 
In reply to a post by MrSaffron:
Two traceroute just done to somewhere in Leeds and another in Luton, so will be interesting to see if things change during peak time period


Hi Andrew,

VM actually have 195.66.224.22, 195.66.224.23, 195.66.225.103, 195.66.236.23, 195.66.237.103 and 195.66.244.33 onto the peering LANs - each of their 6 100GE links is a routed interface.

The routes shouldn't change at peak times - would be noticed on BQM either way.
Standard User jchamier
(eat-sleep-adslguide) Tue 17-Apr-18 23:34:40
Print Post

Re: Thinkbroadband Speedtest capped download speed?


[re: nemeth782] [link to this post]
 
If your requirement is for download speed (eg you use Steam!) then there is no question. If you use obscure websites and corporate (rather than home user) internet services then other ISPs are available.

Download speed is not the only measure of worth of an internet connection. It is a very common one, and one VM rely on.

plusnet unlimited fibre 80/20 - 2 Jun 14 - Sync at 8/Apr/18: 62,153/9,919 - G.INP & 3.0 dB SNRm
19 years of broadband, from 1999's ntl:cable modem trial - Now using Asus RT-AC88U with BT HG612 - BQM

Edited by jchamier (Tue 17-Apr-18 23:35:21)

Pages in this thread: 1 | 2 | 3 | 4 | 5 | [6] | 7 | (show all)   Print Thread

Jump to