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

Login with username, password and session length
Advanced search  

News:

Pages: [1] 2

Author Topic: Ongoing issues with Single Thread Speeds  (Read 5278 times)

jaydub

  • Reg Member
  • ***
  • Posts: 351
Ongoing issues with Single Thread Speeds
« on: December 28, 2017, 10:36:44 PM »

I moved to Fibre with Pulse8 in November 2016 and had single thread speed problems almost immediately.  After some lobbying by Adam at Pulse8 with TTB, the root cause was eventually identified and I had a period of pretty solid single thread speeds.

This deteriorated with the advent of the non-Flash based TTB tester in May this year, which I eventually traced down to my AV, Eset Cyber Security Pro, impacting the single thread results.  this was rectified by swapping to KIS which appears to have a negligible effect on the speed test.

For most of the rest of my time with Pulse8, the single thread speeds were fairly solid although there was always a slight undercurrent of poorer results.  After 12 months I decided to migrate to Aquiss.  The result was very good single thread speeds out of the working day, but their business prioritisation and/or the Entanet node I was connected to running a bit hot, resulted in very variable results throughout the day.

Martin Pitt at Aquiss very graciously made the offer for me to migrate away without penalty and in mid-December I migrated to IDNet on their Fluid Data TTB backhaul.

Since then speeds have been quite variable without their being any obvious pattern to it.  Out of 5 consecutive tests I might get one really good one, one poor one and three middling ones regardless of when in the day I am testing.

I attach a scatter chart for the last three months of my TTB tests.

In general terms the IDNet connection is giving me worse single thread results than either Aquiss or Pulse8 (see figures below).

           IDNet   Aquiss   Pulse8
         
Mean     59.7     60.8     64.2
Median   63.6     69.5     66.9

As the distribution is somewhat skewed the median figures are probably more significant than the mean.

I reliably sync at 80/20 with a 9dB margin and my connection passes the normal line tests that ISPs do (as confirmed by both Aquiss and IDNet).

My MDWS results are available under jaydub and seem fairly unremarkable although my Quiet Line graph is noisier than ideal.

I am in discussions with IDNet about where to go from here, but would welcome your thoughts.

A few questions from my end:
  • Do most people's TBB results have the single thread result coincident with the multi thread test?
  • Is cabinet congestion a possible cause of the single thread degradation, bearing in mind that that the results are pretty much consistent whatever time I test?
  • Is what I am seeing just a factor of my line and unless the line worsens sufficiently for there to be a fault condition, there is very little that can be done about it?
  • IDNet are offering to move the backhaul service from TTB.  Is there any reason not to go ahead with this?
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: Ongoing issues with Single Thread Speeds
« Reply #1 on: December 28, 2017, 11:08:37 PM »

IDNet are offering to move the backhaul service from TTB.

From TTB to whom?
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.

jaydub

  • Reg Member
  • ***
  • Posts: 351
Re: Ongoing issues with Single Thread Speeds
« Reply #2 on: December 28, 2017, 11:32:33 PM »

From TTB to whom?
The last email I had from them suggested BTW, which has confused me as I thought they either used Fluid Data or Zen for backhaul services and consequently BTW only came into play if your cabinet wasn't connected to a Zen PoP, which ours is.

I have sought clarification from IDNet.
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: Ongoing issues with Single Thread Speeds
« Reply #3 on: December 28, 2017, 11:52:23 PM »

I see, thank you. That is a little puzzling.  ???

I think that most of us who take note of "what is going on" will be aware that A&A use both BTW and TTB for backhaul . . . with the latter being slightly more favoured than the former.

Not intending to be provocative but I have to ask the question --

Does the problem with single thread speeds actually stop you making the use of the provided broadband service connecting you to the Internet?
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.

jaydub

  • Reg Member
  • ***
  • Posts: 351
Re: Ongoing issues with Single Thread Speeds
« Reply #4 on: December 29, 2017, 12:16:09 AM »

I see, thank you. That is a little puzzling.  ???

I think that most of us who take note of "what is going on" will be aware that A&A use both BTW and TTB for backhaul . . . with the latter being slightly more favoured than the former.

Not intending to be provocative but I have to ask the question --

Does the problem with single thread speeds actually stop you making the use of the provided broadband service connecting you to the Internet?

Being absolutely honest, with IDnet the answer is no.

But the scientist/engineer in me wants to understand why, if others get single thread speeds that match their multithread speeds within 1Mbps or so, I don't.

And there's the tight fisted Yorkshire git in me, that wonders why I am paying more than I was with Pulse8 to get less speed.

Ultimately there might be an element of 'you live and learn by your mistakes' going on here, but I would like to use this as a learning opportunity to understand where the issues might be and what, if anything, I can do about them.
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: Ongoing issues with Single Thread Speeds
« Reply #5 on: December 29, 2017, 12:30:23 AM »

As a fellow scientist, I can now say that I understand your logic.  :)
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.

jaydub

  • Reg Member
  • ***
  • Posts: 351
Re: Ongoing issues with Single Thread Speeds
« Reply #6 on: December 29, 2017, 12:51:18 AM »

As a fellow scientist, I can now say that I understand your logic.  :)

 ;D
And a little element of a dog with a bone.  ;)
Logged

Chrysalis

  • Content Team
  • Addicted Kitizen
  • *
  • Posts: 7382
  • VM Gig1 - AAISP L2TP
Re: Ongoing issues with Single Thread Speeds
« Reply #7 on: December 29, 2017, 01:03:01 AM »

my single tbb matches the multi yes, if its variable 24/7 regardless of peak hours imo it means one off following

1 transit problem between isp and tbb
2 congestion thats persistent all day, if it were this it would still be noticeably worse at peak so unlikely
3 misconfigured traffic shaping at ISP
4 local problem in your network
Logged

jaydub

  • Reg Member
  • ***
  • Posts: 351
Re: Ongoing issues with Single Thread Speeds
« Reply #8 on: December 29, 2017, 10:47:52 AM »

my single tbb matches the multi yes, if its variable 24/7 regardless of peak hours imo it means one off following

1 transit problem between isp and tbb
2 congestion thats persistent all day, if it were this it would still be noticeably worse at peak so unlikely
3 misconfigured traffic shaping at ISP
4 local problem in your network

Thanks, Chrysalis.

So following this logic through:

1 Unlikely because other IDNet customers get their single thread speeds matching their multi.
2 No obvious time of day pattern with the results.  No sign of congestion on my BQM.
3 Unlikely because other IDNet customers get their single thread speeds matching their multi.
4  All things are possible.  Oldish master socket fitted with Mk3 faceplate filter fitted on Engineer install in November 2016. Testing taking place on an iMac on a dedicated browser with no add ons.  Testing from the master socket with a dangly filter makes no difference.  I've swapped out the router also tried changing Ethernet and modem cables with no apparent difference. 

I would like it to local TBH, as that would be a simple fix.  The only thing I haven't ruled out is that it's the iMac itself causing issues, but at the moment I don't have any other hardware I can directly cable into the router.
Logged

jaydub

  • Reg Member
  • ***
  • Posts: 351
Re: Ongoing issues with Single Thread Speeds
« Reply #9 on: December 29, 2017, 01:37:39 PM »

As a further update:

The only bit of my connection I hadn't swapped out was the iMac itself.

I've just dome some testing from my son's laptop on an ethernet connection to the router and can see no significant difference in the test results.

It was what i expected, but did just want to rule out that it was the Mac causing the issues.

I've also asked a neighbour to run me some results off to see whether he sees anything similar.
Logged

Chrysalis

  • Content Team
  • Addicted Kitizen
  • *
  • Posts: 7382
  • VM Gig1 - AAISP L2TP
Re: Ongoing issues with Single Thread Speeds
« Reply #10 on: December 30, 2017, 06:29:43 AM »

Can you stick in a rescue OS USB in the imac, boot off it, so you can speedtest using a different OS at its defaults?

Various images come with a browser that can be used so can run speedtests etc.
Logged

jaydub

  • Reg Member
  • ***
  • Posts: 351
Re: Ongoing issues with Single Thread Speeds
« Reply #11 on: December 31, 2017, 12:10:39 AM »

Can you stick in a rescue OS USB in the imac, boot off it, so you can speedtest using a different OS at its defaults?

Various images come with a browser that can be used so can run speedtests etc.
Have struggled with creating a rescue disk off High Sierra, but have created a bootable install disk and reloaded the OS from there.

No noticeable difference, as expected since a Win 10 laptop gives similar results.
Logged

jaydub

  • Reg Member
  • ***
  • Posts: 351
Re: Ongoing issues with Single Thread Speeds
« Reply #12 on: January 03, 2018, 08:38:49 PM »

IDNet are going to change the backhaul from TTB to BTW in an effort to isolate the LLU equipment at the exchange being the source of the issue.  I seem to have done enough to convince them that it isn't a local issue.

Having had fibre on two TTB backhaul providers (Pulse8 and IDNet) and one BTW (Aquiss) over the last few months, Aquiss were the only ones that provided consistent single thread download speeds (albeit only out of business prioritisation hours), so this may work.

If consulted, I'd have rather tried Zen, but I understand their logic.  Time will tell whether it is right or not.  The backhaul transfer takes place on Jan 16th.

After that, I think I'll either feel vindicated or think I ought put up and shut up.
Logged

PhilipD

  • Reg Member
  • ***
  • Posts: 591
Re: Ongoing issues with Single Thread Speeds
« Reply #13 on: January 06, 2018, 12:06:54 PM »

Hi

Just to add I've not had maximum throughput via Uno (which use TalkTalk) for many, many months on the single thread speed test, regardless of the time day or night, I usually get around 60-64Mbps on the single thread test, with the multi-thread test always ~74Mbps. I know when you first posted about this and I replied I was getting maximum on both types of test, but at some point that changed, although nothing my end has changed.

However playing about with TCP Optimizer https://www.speedguide.net/downloads.php I can improve the single thread speeds back up to almost match the the multi-thread speed, so I don't think it is something being deliberately throttled by TalkTalk, just some interaction between Windows (10 in my case) automatic network settings just not hitting an optimal match against TalkTalk and the site it's connecting to.  This may explain why some speed tests are okay and others are not.

Edit:  if I test here https://testmy.net/results and use the single thread speed test option I get consistently > 73Mbps, it's only ThinkBroadband's single threaded part of the test that is impacted, which I think was your original issue?  I think the problem is with Thinkbroadband and how the test is routed or network parameters and there isn't really any problem at all with the speeds, just the test.
 

Regards

Phil




« Last Edit: January 06, 2018, 12:54:02 PM by PhilipD »
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Ongoing issues with Single Thread Speeds
« Reply #14 on: January 06, 2018, 12:20:20 PM »

Didn't AA have a similar experience with TT at some point? I can't remember.

Would AA suit you? They are very keen on performance monitoring.
Logged
Pages: [1] 2
 

anything