Re: Bufferbloat related censorship at Virgin Media
On Sun, Mar 1, 2015 at 3:47 PM, Mel Beckman m...@beckman.org wrote: Dave, I appreciate all your work on buffer bloat. It looks like you have done quite a lot of selfless contribution. However, I don't think you're effectively communicating with the people who can change things. After I read what you said, here is what I would have heard as a service provider: I am the smartest person in the room. I am pretty close to being the smartest person in the room. That said, people like van jacobson, eric dumazet, tom herbert, jim gettys, eric raymond, vint cerf, dave reed, fred baker, and many, many others, smarter than me, have also been banging this drum, politely, and rationally, to not much effect, for 4+ years now. google for any of those names and the word bufferbloat. You better listen to me, because if you don't there will be trouble. But you probably won't because you're too stupid. Your customers suffer because you are idiots. Listen to me! This issue is too important for me to be polite, or even coherent. If you can't figure out what I'm saying, do some research and figure it out! Plus, apologize to me! I demand it! Oh, banning my ip for *3 links to sane benchmarks and fixes*, really pushed me over the edge. Bees, honey, vinegar, etc. I have been polite, constructive, and helpful, for four+ years. I have worked both in the background and foreground with many companies, to start hopefully, getting bufferbloat fixed across the entire edge of the internet. It hasn't worked fast enough for my liking, and the last batch of new products that claimed to fix it, didn't, and the market is now rife with genuine lies as to whether they did or not. So, this morning, I tried this. Sorry for the noise on these lists. Honestly! I totally agree with your assessment of my tone, btw! but I would rather like the cable industry in particular, to come clean, with schedules for deployable fixes. I am off to go fix wifi next, and I do hope that 2+ billion people in the world - if not the isps, maybe - would like wifi to get better also, and indeed, I spent the weekend constructively starting to implement some of the fixes I outlined at the last 802.11 meeting I attended. That part, on fixing wifi bufferbloat - a much harder problem than edge bufferbloat - , is a lot of fun! For some info on what we plan to do there, see: http://snapon.lab.bufferbloat.net/~d/ieee802.11-sept-17-2014/11-14-1265-00-0wng-More-on-Bufferbloat.pdf So I took a break from that, reared back, and got some stuff off my chest. -- Dave Täht Let's make wifi fast, less jittery and reliable again! https://plus.google.com/u/0/107942175615993706558/posts/TVX3o84jjmb
Re: Bufferbloat related censorship at Virgin Media
On Sun, Mar 1, 2015 at 4:04 PM, Jack Bates jba...@paradoxnetworks.net wrote: On 3/1/2015 5:28 PM, Dave Taht wrote: My IP address is apparently now banned from accessing your site at all, for advertising, on this thread: http://community.virginmedia.com/t5/Up-to-152Mb/Bufferbloat-High-Latency-amp-packet-loss-when-connection/td-p/2773495 I don't see how codel is related to the customer complaint from their perspective. The problem appears to be high latency on downstream with little to no upstream. I'd probably call it off-topic advertising. The only thing that seems to relate to codel is the user's use of bufferbloat in the topic. Nothing the user can do will fix the downstream to my knowledge. Not that I'm extremely knowledgeable on the subject. It is 100% possible to fix excessive downstream buffering from some misconfigured device with a shaper on the download *on the CPE or home router*. I have been doing that for 15 years. So has everyone that uses nearly any of the shapers that are available for Linux, at least. http://burntchrome.blogspot.com/2014_05_01_archive.html doing it yourself, right, requires a good measurement, and you lose just a little bit of single-flow bandwidth - typically 5% - but you get it all back with faster tcp ramp up times, huge improvements in dns lookups, voip, gaming, and other traffic. it generally works way better than policers do. Jack -- Dave Täht Let's make wifi fast, less jittery and reliable again! https://plus.google.com/u/0/107942175615993706558/posts/TVX3o84jjmb
Re: Bufferbloat related censorship at Virgin Media
On 3/1/2015 6:14 PM, Dave Taht wrote: It is 100% possible to fix excessive downstream buffering from some misconfigured device with a shaper on the download *on the CPE or home router*. From OP: However I've recently noticed periods of 500-800ms latency to the CMTS gateway when only using 15-20 of the 60Mbps total (and little to none upstream utilisation). I agree with you that it is better to run a shaper that insures your shaper hits saturation and handles queue policies before the upstream does. That is great if it is your pipe (and only its queue) that is saturating. I don't think this problem qualifies. I find it difficult to believe that he's hitting a buffer bloat issue on a single (not shared with others) queue using 1/3rd of the total bandwidth available to him at those speeds and with that latency value. His problem is more likely lower down (unable to obtain max speed resulting in saturation) or a shared queue where others are saturating it and him applying a shaper will not keep others from doing so. Jack
Re: Bufferbloat related censorship at Virgin Media
Well, with luck probably it will just bounce off their corporate hull and drift into the Kuiper belt. Say hi to Sugar ;) -mel On Mar 1, 2015, at 4:01 PM, Dave Taht dave.t...@gmail.com wrote: On Sun, Mar 1, 2015 at 3:47 PM, Mel Beckman m...@beckman.org wrote: Dave, I appreciate all your work on buffer bloat. It looks like you have done quite a lot of selfless contribution. However, I don't think you're effectively communicating with the people who can change things. After I read what you said, here is what I would have heard as a service provider: I am the smartest person in the room. I am pretty close to being the smartest person in the room. That said, people like van jacobson, eric dumazet, tom herbert, jim gettys, eric raymond, vint cerf, dave reed, fred baker, and many, many others, smarter than me, have also been banging this drum, politely, and rationally, to not much effect, for 4+ years now. google for any of those names and the word bufferbloat. You better listen to me, because if you don't there will be trouble. But you probably won't because you're too stupid. Your customers suffer because you are idiots. Listen to me! This issue is too important for me to be polite, or even coherent. If you can't figure out what I'm saying, do some research and figure it out! Plus, apologize to me! I demand it! Oh, banning my ip for *3 links to sane benchmarks and fixes*, really pushed me over the edge. Bees, honey, vinegar, etc. I have been polite, constructive, and helpful, for four+ years. I have worked both in the background and foreground with many companies, to start hopefully, getting bufferbloat fixed across the entire edge of the internet. It hasn't worked fast enough for my liking, and the last batch of new products that claimed to fix it, didn't, and the market is now rife with genuine lies as to whether they did or not. So, this morning, I tried this. Sorry for the noise on these lists. Honestly! I totally agree with your assessment of my tone, btw! but I would rather like the cable industry in particular, to come clean, with schedules for deployable fixes. I am off to go fix wifi next, and I do hope that 2+ billion people in the world - if not the isps, maybe - would like wifi to get better also, and indeed, I spent the weekend constructively starting to implement some of the fixes I outlined at the last 802.11 meeting I attended. That part, on fixing wifi bufferbloat - a much harder problem than edge bufferbloat - , is a lot of fun! For some info on what we plan to do there, see: http://snapon.lab.bufferbloat.net/~d/ieee802.11-sept-17-2014/11-14-1265-00-0wng-More-on-Bufferbloat.pdf So I took a break from that, reared back, and got some stuff off my chest. -- Dave Täht Let's make wifi fast, less jittery and reliable again! https://plus.google.com/u/0/107942175615993706558/posts/TVX3o84jjmb
Re: Bufferbloat related censorship at Virgin Media
On 3/1/2015 5:28 PM, Dave Taht wrote: My IP address is apparently now banned from accessing your site at all, for advertising, on this thread: http://community.virginmedia.com/t5/Up-to-152Mb/Bufferbloat-High-Latency-amp-packet-loss-when-connection/td-p/2773495 I don't see how codel is related to the customer complaint from their perspective. The problem appears to be high latency on downstream with little to no upstream. I'd probably call it off-topic advertising. The only thing that seems to relate to codel is the user's use of bufferbloat in the topic. Nothing the user can do will fix the downstream to my knowledge. Not that I'm extremely knowledgeable on the subject. Jack
Re: Bufferbloat related censorship at Virgin Media
http://burntchrome.blogspot.com/2014/05/disabling-shaping-in-one-direction-with.html On Sun, Mar 1, 2015 at 4:39 PM, Jack Bates jba...@paradoxnetworks.net wrote: On 3/1/2015 6:14 PM, Dave Taht wrote: It is 100% possible to fix excessive downstream buffering from some misconfigured device with a shaper on the download *on the CPE or home router*. From OP: However I've recently noticed periods of 500-800ms latency to the CMTS gateway when only using 15-20 of the 60Mbps total (and little to none upstream utilisation). Might be. Again, all I did on that thread was provide a few pointers to bufferbloat related resources, and pointed at the downlink being a real problem quite often with links to stuff like this http://burntchrome.blogspot.com/2014/05/disabling-shaping-in-one-direction-with.html and they yanked me. Certainly I could have tried again, from another IP, but ya know, some sundays are more fun than others. I agree with you that it is better to run a shaper that insures your shaper hits saturation and handles queue policies before the upstream does. That is great if it is your pipe (and only its queue) that is saturating. I don't think this problem qualifies. Might not. That said, it was hardly an accurate measurement. It is also perfectly feasible for the upstream device or the downstream device to be measuring these problems and deal with them appropriately. It gets progressively easier cpu-wise, as the effective bandwidth goes down. It is unfortunately nearly impossible for the next device in line to do (although we have some tools measuring interpacket smoothness that can provide a hint now, they are not baked yet) It was my hope, in working on the DOCSIS 3.1 standard that all the possible downstream problems would be addressed. They weren't. I find it difficult to believe that he's hitting a buffer bloat issue on a single (not shared with others) queue using 1/3rd of the total bandwidth available to him at those speeds and with that latency value. His problem is more likely lower down (unable to obtain max speed resulting in saturation) or a shared queue where others are saturating it and him applying a shaper will not keep others from doing so. Jack -- Dave Täht Let's make wifi fast, less jittery and reliable again! https://plus.google.com/u/0/107942175615993706558/posts/TVX3o84jjmb