07-11-2021 20:16 - edited 07-11-2021 20:25
Having suffered with poor upstream latency under load (aka 'bufferbloat') for years, as have VM's other customers, I was interested to read a paper today by Flickinger et al.
The paper refers to a newer upstream active queue management, Proportional Integral Controller Enhanced - known colloquially as DOCSIS-PIE. More specifically, the paper discusses the impact of implementing DOCSIS-PIE on a (Comcast) DOCSIS network during the COVID-19 pandemic, when everyone started working and schooling from home, thus adding unprecedented latency and load to networks.
A rollout by Comcast with two modems, one with AQM/PIE enabled and one without, showed a difference in upstream latency under load to the order of 15ms (with PIE) vs 250ms (without PIE). A rather staggering difference! I certainly don't see behaviour as good as that on VM, hence the thread/question.
More importantly, the paper goes on to cite that:
In 2014, CableLabs included PIE as the official AQM in the DOCSIS 3.1 specifications as a mandatory feature for the CMTS and cable modems (PIE was also added as an optional feature for earlier generation DOCSIS 3.0 cable modems). At the same time, CableLabs required that DOCSIS 3.1 CMTS equipment support AQM as well but left the algorithm choice to the vendor.
Given it's mandatory in the standard/specification, I'm assuming this is something that VM has enabled for the downstream DOCSIS 3.1 channel on Gig1? What of the other DOCSIS 3 channels, which are optional?
(Edit: I'm asking about the CMTS and the CPE here.)
There's scant information out there on the VM network core, and this is the only place I know of to ask such a question despite the fact the forum is clearly not geared to this type of discussion (regrettably).
I'd be interested to hear a definitive reply back from someone at VM, and I do understand that the forum team will likely need to bounce this over to someone at networks for comment.
It'd be nice to finally have a connection that doesn't collapse into a crippled mess of high latency because someone's phone is backing up to iCloud, when I'm trying to (for example) make a video call or waiting for elements to render on a website. As it is, currently I'm just implementing fq_codel on my OpenBSD router (and running cake+bbr on my *nix boxes). But what of VM's end?
 Improving Latency with Active Queue Management (AQM) During COVID-19
By Allen Flickinger, Carl Klatsky, Atahualpa Ledesma, Jason Livingood, Sebnem Ozer
on 17-11-2021 20:45
It looks as though our community have this in hand.
Hardly this is for someone higher up in networks for docsis
on 18-11-2021 12:07
It looks as though our community have this in hand. Is there anything that you wish us to look into?
Really? With respect, only (off the top of my head) one other member even understood the question. As per the OP, I want to know whether VM implement DOCSIS-PIE; and if not, why not (and when)? Bufferbloat has blighted the network for years, and it's (relatively) easily fixed. The studies I produced with Dave Täht, Vint Cerf (the co-inventor of the Internet) et al. and Comcast showed massive benefits to end users during the pandemic, and improved load on the network itself. Maybe you could bounce the question (and flent/netperf graphs) up to someone in Networks and ask them for a comment?
on 09-10-2022 11:14
on 11-10-2022 14:29
Hey tommy40, thank you for reaching out and a warm welcome to the community, I am sorry to hear you are having some connection issues.
I have taken a look at our end and I can't see any issues, the levels are really good.
What type of issues are you having, is it on WIFI or cable?
Please do let me know. Thanks
Matt - Forum Team
New around here?
on 11-10-2022 16:57
Buffer-blot on cable
28-10-2022 00:30 - edited 28-10-2022 00:34
Sorry for the delay replying, I'm not a regular here. Bufferbloat will always be an issue on cable (DOCSIS). Unfortunately VM seem to have no interest in PIE or even discussing it. See how long this thread has been running, with no staff reply, for details...
You need to put your Hub into modem mode and install a router with PIE (or at least a decent fq_codel implementation) to help alleviate bufferbloat. Ideally, an x86 box running something like OpenWRT or OPNSense. Commercial routers which have working SQM at gigabit speeds are limited, but apparently the Eero mesh routers (PIE) and some Ubiquiti routers (fq_codel) can manage it.
How are you measuring the bufferbloat? Are you using a site like WaveForm or DSLReports, or have you run the Flent RRUL test to measure 'properly'? Check out bufferbloat.net (Dave Täht's site) and flent.org for details.