Many vdsl connections send data over the interleave path without interleaving being applied.
The FAST path may be reserved for certain types of time critical traffic such as IPTV.
All other general data is then sent over the Interleaved path. Interleaving can be turned off on the Interleave path by using interleave depth of 1. This is a DLM configuration and not something you can control nor over ride as its set on the MSAN.
There isn't anything you can configure that will improve your latency. As mentioned first hop depends a lot on your location and many of the hops are hidden.
Some BTr connections show the RAS as first hop. Other ISPs may purchase a product where L2TP hides all the Openreach hops and shows the ISP gateway as being first hop.
Approx 8/9 years ago BTOR upgraded the 21CN network to do away with the traditional BT RAS which were located in about 8 core locations and installed
MSE bRAS putting the bRAS nearer to home. Many are in or near to what may be your headend exchange. There's approx 500 MSE bRAS locations compared to just 8 old BTw RAS.
This makes a massive different to BTr customers who see the BTw RAS as being local. As mentioned in my earlier post, it takes on average approx 6ms to get to your local exchange and on to the BTw backhauls. Their IPs are dished out by the local bRAS pool and they can actually see the MSE bRAS as their first hop.
ISPs who have purchased other products, say
shared WBMC deliberately have the (BTw bRAS) hidden and even though their customers will go through that same MSE bRAS, the first hop that shows is the ISP gateway usually down in one of the London CoLos. As per prev post it can take 2-3ms on the backhaul/core down to London depending up;on your location.
You could have 2 people in Bolton. One with BTretail, the other with Noddy ISP
The BTr user tracert will show something like:
- 1ms (router)
- 6ms (Bolton MSE)
- 8ms (Core Location)
- 9ms (jumping around various London nodes/interconnects/peering/transit)
The person with Noddy ISP will see something like:
- 1ms (router)
- 8ms (ISP gateway - usually London)
- 9ms (jumping to various gateways/interconnects/peers/transit)
Both connections may have the same routing up until the ISP gateway. The difference is BTr connections show the RAS hop. ISPs usually always tend to hide the BTw RAS hop. Although it may look like BTr has lower first hop latency, this is only because the 1st hop is likely your head end exhange whereas other ISPs tunnel traffic up to their gateway in London.
I don’t think I am on VDSL2 fast path right now but interleaving via G.998.4.
G.INP has additional bearer channels. Any interleave settings on Bearer 1 has no effect on your data to the internet on Bearer 0.
Those settings are for retransmission in the case of noise events. I don't think you can change them... but anyhow leave them well alone and dont even attempt tp touch. If you break those settings then you could well end up with a pile of additional latency on both bearers.
The connection starts off around 4.x MS and then increases to 8ms. Then it should be possible to some how keep the lower latency,
I've just this min noticed you don't appear to be in the UK, but the theory is the same. In the UK 4-6ms is usually a hop at or near to your headend exchange. It then increases to about 8ms to get to the interconnects in London, but possibly Melbourne for you. Most of the early steps of latency are due to distance.
This is first hop latency from the modem and not the total latency on a path.
Just checking what you call first hop and to ensure we're not talking about different things. Are you saying first hop to the modem/router. ie the very first hop that is usually 1ms if you use Ethernet. This figure can be higher if you use wireless.
From what i have read up it is related to interleaving and G.INP.
Interleaving does affect latency. Being on FAST or INTERLEAVE path makes no difference if the Interleave depth is 1. Depth = 1 means there's no interleaving on the interleaved path. Most VDSL2 connections use the interleave path for normal data so that the fast path can be reserved for time critical apps such as voice and iptv. This means the ISP has the flexibility where interleave can be controlled by DLM and switched on/of as required for normal data.
using "xdslctl configure --maxDataRate"
That doesnt have anything to do with interleave settings. Its capping the max speed of your line. There may be some confusion in what you have read - ie we sometimes use this if the Openreach DLM has got stuck and to kick start the DLM causing it to see a sync speed with a large surplus SNRm. It's a trick to wake up the Openreach DLM if your line is interleaved. However the setting itself doesnt change anything other than cap the line rate and causing a large SNRm.
I believe this will be related to either "0x00004000 CfgFlagsDisableVectoring"
Here in the UK very few lines VDSL2 lines have vectoring. Its mostly used with G.fast connections. I think similar applies in Aus. Even if you did have vectoring, you wouldnt want to disable it. Vectoring helps cancel noise meaning less need for Error Correction and high levels of Error Protection. Its not related to maxDataRate.
forcing max depth to 1 just not sure on the possible commands.
As above, this is a DLM setting configured by the SP on the DSLAM. Its not something set modem side.
Wait!! I've just noticed the stats on the compare link
Max: Upstream rate = 50282 Kbps, Downstream rate = 145234 Kbps
Bearer: 0, Upstream rate = 44199 Kbps, Downstream rate = 115831 Kbps
VDSL2 Profile: Profile 17a
I'm now confused what product you are on. That's not standard vdsl2 80/20.
D: 16 4
That shows Bearer 0 has interleave depth of 16
The first set of stats has SRA. Do you have g.fast?
SRA: 4/4 0/0