On Fri, Mar 31, 2017 at 7:30 PM, Jim Gettys <[email protected]> wrote:
> > > On Fri, Mar 31, 2017 at 6:18 PM, Joe Touch <[email protected]> wrote: > >> >> >> On 3/31/2017 7:47 AM, Jim Gettys wrote: >> > One full size packet @ 1mbps == 13 milliseconds. >> That sort of blocking is its own problem and the only solution is to use >> smaller packets. >> >> It can be solved at TCP by pushing the MSS down, but for UDP that's >> another good reason for considering UDP options as a TSVWG WG item! >> > > Yes, but without better packet scheduling such as fq_codel at the > bottleneck, it won't help. But I'm decently optimistic that that > fq_codel/cake/other things will be deploying widely, since we're seeing > fq_codel appear in commercial products at this date. > Oh, I forgot: I wrote this piece recently. It may be useful to people, and let me know if there are products you think should be added to the list. https://gettys.wordpress.com/2017/02/02/home-products-that-fixmitigate-bufferbloat/ The Evenroute router is particularly good if you need a "plug and play" home router, and/or one that can deal with ISP's that under provision their network and deliver variable bandwidth, such as my brother's DSL line. I need to check if the latest Evenroute supports the new WiFi code; that is stunningly good. My Ubiquiti Picostation running LEDE keeps latency around 40ms or less even under load :-). I've just installed a DOCSIS 3.1 modem yesterday, and will probably run some tests on it versus what SQM does in the next week. That should be interesting. It's particularly frustrating that major vendors "bait and switch" their customers with trade names. Said vendor(s) will get no endorsement from me, despite the fact it's enough of the market to be "interesting". Would be nice to get pacing into the new WiFi code and fq_codel so we can use BBR with it on clients (Linux/Android). Any volunteers for an adventure? - Jim > > >> Joe >> >> _______________________________________________ >> iccrg mailing list >> [email protected] >> https://www.irtf.org/mailman/listinfo/iccrg >> > >
