Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Author Topic: PPPoEoA  (Read 542 times)

Weaver

  • Kitizen
  • ****
  • Posts: 4004
  • Retd sw dev; A&A; 3 × 7km ADSL2; IPv6; Firebrick
PPPoEoA
« on: April 06, 2017, 07:43:10 AM »

Is anyone out there apart from me using PPPoE over ATM on the DSL link?

As I mentioned earlier, I am intending to change my modems’ config over from using PPPoEoA with RFC2684 LLC to use PPPoEoA RFC2684 VC-MUX. If it works then I am expecting a 3.125% speed improvement on full length 1500 byte IP packets, which comes from a total of 33 ATM cells being reduced to 32. See
    https://wand.net.nz/~513/2006/readings/adsl-2.pdf

Check my logic, but VDSL2 users and Draytek Vigor users are ruled out.
Logged

d2d4j

  • Reg Member
  • ***
  • Posts: 422
Re: PPPoEoA
« Reply #1 on: April 06, 2017, 08:14:56 AM »

Hi weaver

I believe we have a few clients clients who run adsl2+ on vc-mux

What is your question

Our clients connections may not be appropriate though, due to very short lines directly into the telephone exchange itself and speeds in excess of 20mb down and they are business connections

Many thanks

John
Logged

Weaver

  • Kitizen
  • ****
  • Posts: 4004
  • Retd sw dev; A&A; 3 × 7km ADSL2; IPv6; Firebrick
Re: PPPoEoA
« Reply #2 on: April 06, 2017, 11:37:37 PM »

It works.  ;D

I don't remember why I was convinced that PPPoEoA VC-MUX doesn't work (on BT). Perhaps it doesn't work on 20CN?

A pleasing speed improvement anyway, with the highest ever performance figures reported by speedof.me 7.9 Mbps down, 1.37 Mbps up. I suspect that speedof.me exaggerates by 4 - 10%, but I am comparing like with like. Ookla reports 7.27 / 1.21 best cases.

Logged

Weaver

  • Kitizen
  • ****
  • Posts: 4004
  • Retd sw dev; A&A; 3 × 7km ADSL2; IPv6; Firebrick
Re: PPPoEoA
« Reply #3 on: April 07, 2017, 12:22:21 AM »

In terms of efficiency, calculated as ( speed_tester / sync ), there seems to be a 3% improvement in downstream efficiency, exactly as predicted, and a 5% improvement upstream which is a bit surprising, although these are rather rough figures. These were based on the highest speed tester values only, taken from speedof.me, not on the mean. Although speedof.me does exaggerate, I feel sure, I am comparing like with like and the comparison is based on the ratio of efficiency values not on absolute differences.

So every little helps, it works for BTW 21CN ADSL2, and if it works for you the definitely go for it. It should always be the first choice. Not applicable of course if you are on VDSL2 as there's no cursed ATM, no PPPoEoA and therefore no inefficiency to worry about so nothing to improve.
« Last Edit: April 07, 2017, 12:25:48 AM by Weaver »
Logged

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 18960
  • Over the Rainbow
    • The ELRepo Project
Re: PPPoEoA
« Reply #4 on: April 07, 2017, 01:53:22 AM »

It works.  ;D

  :yay:

Quote
I don't remember why I was convinced that PPPoEoA VC-MUX doesn't work (on BT). Perhaps it doesn't work on 20CN?

I seem to recall reading a definitive statement, somewhere, that wherever "PPPoA" is written it is, in fact, "PPPoEoA" . . . apparently the "oE" is left out for brevity.

PPPoA (as I always type), VC-MUX 0/38, worked for TT on 20CN and works for TT on 21CN. I am unable to make any statement with regards to BT.
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

j0hn

  • Reg Member
  • ***
  • Posts: 663
Re: PPPoEoA
« Reply #5 on: April 07, 2017, 02:16:25 AM »

It was confirmed in a similar thread that PPPoEoA VC-MUX also works on BTw 20CN.
Logged
BT FTTC 55/10 ECI Huawei Cab - Zyxel VMG8924 bridge mode + Asus RT-AC68U - minted on MDWS via DslStats

Weaver

  • Kitizen
  • ****
  • Posts: 4004
  • Retd sw dev; A&A; 3 × 7km ADSL2; IPv6; Firebrick
Re: PPPoEoA
« Reply #6 on: April 07, 2017, 03:08:01 AM »

I think I picked up the PPPoEoA terminology from Cisco papers and similar.

When I use the term PPPoA I understand it to mean something completely different - Rfc2364 ie PPP with VC-MUX over AAL5 over ATM over ADSL and no Ethernet MAC headers at all on the DSL link. Isn't that what people generally mean by PPPoA for example when they are using G.992.x modem-routers?
Logged

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 18960
  • Over the Rainbow
    • The ELRepo Project
Re: PPPoEoA
« Reply #7 on: April 07, 2017, 08:06:03 PM »

I prefer not to say . . . as my understanding is minimal. I find it like trying to see to the bottom of a pond after the local hogs have been washed therein!  :D
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

ejs

  • Kitizen
  • ****
  • Posts: 1261
Re: PPPoEoA
« Reply #8 on: April 07, 2017, 08:16:47 PM »

I seem to recall reading a definitive statement, somewhere, that wherever "PPPoA" is written it is, in fact, "PPPoEoA" . . . apparently the "oE" is left out for brevity.

I think it's more likely that the statement was something about PPPoE when used on ADSL is really PPPoEoA, with the oA usually not written.
Logged

Weaver

  • Kitizen
  • ****
  • Posts: 4004
  • Retd sw dev; A&A; 3 × 7km ADSL2; IPv6; Firebrick
Re: PPPoEoA
« Reply #9 on: April 07, 2017, 08:42:00 PM »

Ejs' interpretation makes sense. I missed it.

I always find it useful to keep the terms PPPoE and PPPoEoA distinct though because of  ashes such as when talking about something like the Draytek Vigor, since there PPPoE on an Ethernet LAN does not imply to PPPoEoA. Also PPPoEoA is hardly meaningfully 'over ethernet' if you are actually talking about a protocol that is running over a DSL link, where the Ethernet MAC headers are sometimes just unnecessary junk which is never even used, as in my case where they are generated in order to minimise performance and then just discarded at the receiving end. I find it helpful to make the distinction between the 6 byte PPPoE header only, which is common to both PPPoEoE and PPPoEoA, versus the entire protocol stack that is added on top of AAL5 in PPPoEoA that is PPPoE + MAC + RFC2684, also optionally including PPP itself in that lot.
Logged

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 18960
  • Over the Rainbow
    • The ELRepo Project
Re: PPPoEoA
« Reply #10 on: April 07, 2017, 09:18:27 PM »

I think it's more likely that the statement was something about PPPoE when used on ADSL is really PPPoEoA, with the oA usually not written.

Ah, that could be what was lurking in the back of my mind. Thank you.

I was debating (with myself (and loosing)) that when PPPoX is stated, the PPP is the "top layer" and the "X" is the "bottom most layer", with all the "intervening layers" left unmentioned. (Whatever "top", "bottom" and "intervening" may mean.)
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

tickmike

  • Kitizen
  • ****
  • Posts: 3010
  • Yes Another Penguin !. :)
    • Free Download from.
Re: PPPoEoA
« Reply #11 on: April 07, 2017, 10:36:51 PM »

Once my line comes out of 'Training' after  isp did a DLM reset, I am going to try my HG612 modem again and try this mode.
Logged
I RECOMMEND TRYING / USING PCLinuxOS (www.pclinuxos.com) .
I have a set of 8 fixed IP's From my Eclipse isp.
BT ADSL2 line>HG612 set as a Modem, Bridge, WAN not Bound to LAN1 or 2 >Smoothwall (Hardware Firewall and routing) > Ethernet LAN, DMZ,WiFI LAN and Spare LAN .
DSLstats LAN2  linked Ethernet

Weaver

  • Kitizen
  • ****
  • Posts: 4004
  • Retd sw dev; A&A; 3 × 7km ADSL2; IPv6; Firebrick
Re: PPPoEoA
« Reply #12 on: April 07, 2017, 11:14:14 PM »

Tickmike, good for you. Get some free speed. There's no need to wait, is there?
Logged

Weaver

  • Kitizen
  • ****
  • Posts: 4004
  • Retd sw dev; A&A; 3 × 7km ADSL2; IPv6; Firebrick
Re: PPPoEoA
« Reply #13 on: April 07, 2017, 11:25:55 PM »

I am going to have to do a lot more tests as the statistics on the results for protocol efficiency are not tight enough. I should be collecting ookla measurements instead of speedof.me as they have a better standard deviation, but I have a lot more speedof.me data and it's ratios of efficiency figures that are being scrutinised so the exaggeration in speedof.me doesn't matter.

It's interesting that the upstream protocol efficiency ratio doesn't match the downstream (upstream efficiency ratio is something like 5% not 3%). This could just be an effect of the limited sample size, I'd have to check, my statistics knowledge is an embarrassment, but if it is significant then it presumably means that upstream is not always using full-size PDUs, and short PDUs would show a far greater efficiency improvement, indeed it could be some thing nearer to 100% than 3.125% as the effect of adding one extra wasteful ATM cell in ratio terms could be enormous for eg TCP ACKs or VoIP. However seeing as this is based on speedtesters’ results, I would have thought it would be dominated by full-size packets. So I don't understand the results at all. Maybe the difference is just a red herring.
Logged