Oh yes, wondered why the IP address was different, kept staring at it, still didn't spot the obvious!

appreciate it as Iike to boast about my ping in blops2 etc

that looks to me odd.

it seems plusnet have their gateways located close to london as is 1ms between your gateway and the bbc. or you are located in london?

Looks like -2ms to me?

6    14 ms    11 ms    13 ms  ae1.ptw-cr01.plus.net [195.166.129.0]
7    12 ms    12 ms    11 ms  kingston-gw.thdo.bbc.co.uk [212.58.239.6]

On my FTTC line I've noticed Plusnet pings go up by about 2-3ms. It coincided with the upgrades to their gateways and traffic management upgrades.

In reply to a post by Chrysalis:

it seems plusnet have their gateways located close to london as is 1ms between your gateway and the bbc. or you are located in london?

Plusnet's gateways are indeed all in London - with feeds off 3 different BT exchanges (Colindale, Faraday, Stepney, all visible on the network performance graphs)

The gateways are, I believe located in 3 different London locations:
ptw - Telehouse West
ptn - Telehouse North
pcl - City Lifeline

I think this diagram is fairly current.

The location of the gateway is seen in the first non-local hops of the traceroute. In my case, the connection is pcl-ag03:

In reply to a post by kasg:

1 <1 ms <1 ms 1 ms zyxel.com [192.168.1.2]
2 13 ms 13 ms 13 ms lo0-central10.pcl-ag03.plus.net [195.166.128.184]

However, it does have to be said that this particular traceroute was done from a location in Surrey, so not too far from London either.

ok thanks, so yeah its different to how bt operates its own service then.

eg. here is my trace, notice a low latency to my gateway. but then higher overall latency than yours. is always a big jump to the ealing hop.

1 <1 ms <1 ms <1 ms home.gateway [192.168.1.253]
2 5 ms 5 ms 5 ms bt.gateway [217.32.144.166]
3 6 ms 6 ms 5 ms bt.internal [217.32.144.190]
4 8 ms 8 ms 7 ms bt.internal [213.120.181.58]
5 7 ms 7 ms 7 ms 217.41.169.213
6 9 ms 9 ms 8 ms bt.internal [217.41.169.109]
7 8 ms 7 ms 7 ms 109.159.251.235
8 16 ms 16 ms 15 ms core2-te-0-13-0-11.ealing.ukcore.bt.net [109.159
.251.173]
9 15 ms 16 ms 14 ms peer2-xe9-1-0.telehouse.ukcore.bt.net [109.159.2
54.114]
10 18 ms 18 ms 16 ms 194.74.65.42
11 * * * Request timed out.
12 * * * Request timed out.
13 15 ms 15 ms 17 ms ae0.er01.telhc.bbc.co.uk [132.185.254.109]
14 18 ms 18 ms 18 ms 132.185.255.140
15 16 ms 16 ms 16 ms www-vip.telhc.bbc.co.uk [212.58.251.195]

That's the experience here, too, when running a Huawei DSLAM on the bench.

The ping response from the PPP "gateway" (surely we should call this an aggregator or access concentrator?) was a minimum of 4ms. That was with the DSL connection placed on a fast path with Forward Error Correction disabled.

The ping time was measured with the "gateway" server connected directly to the DSLAM upstream port by a two metres gbE patch cable. So no issues of geographical distance adding latency. Just the packet processing overheads of the DSLAM itself, and the fast PPP server.

cheers, a

Edited by deleted (Wed 19-Jun-13 21:12:15)

I wonder if the differences are explained by BT running their service as pure WBC, while Plusnet are using a WBMC infrastructure.

All those "bt.gateway" and "bt.internal" timings would then be part of their (BT Retail) WBC infrastructure, in passing back from the individual exchanges to the "ukcore" section.

Meanwhile for Plusnet, the equivalent WBMC infrastructure (owned by BT Wholesale) stays invisible to the IP layer, so gets no individual ping timing until passing back into Plusnet's own infrastructure.

all those ips at the start have no rdns, I set the hostnames in my hosts file to stop tracert taking an age to progress. Ocassionally tho as you can see new ip's popup that I havent got in my hosts file.

But yeah basically between you and telehouse you dont see the hops on plusnet, whilst I do on BT.

Edited by Chrysalis (Thu 20-Jun-13 00:53:16)