Discussion:
vendor status-s for bufferbloat solutions
(too old to reply)
Dave Taht
2018-02-08 20:10:09 UTC
Permalink
It would be nice to have a set of official statements from vendors
regarding their bufferbloat solutions vis a vis the various RFCs
(notably rfc8290) before the ietf blog entry goes out.

My interchange with the tomahawk 3 guy was fruitless. Apple's kind
of exited the router market. Cable manufacturers are obligated to
implement docsis-pie in 3.1. Microsoft has made a big push recently to
modernize their TCPs. A bunch of router makers did embrace fq_codel
(calling it various things like streamboost, adaptive qos, etc), and a
few calling it what it is notably evenroute, and ubnt in their
edgerouter products, sophos, all of lede/openwrt/asuswrt/etc... but
I'm still unaware of any implementations on ONTs, CMTSes, DSLAMs, DSL
firmwares, MOCA...

What I kind of envision is everybody here annoying their favorite
forums for their favorite product and asking what of the outputs of
the ietf aqm working group they planned to implement.
--
Dave Täht
CEO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-669-226-2619
Michael Richardson
2018-02-08 21:53:47 UTC
Permalink
Post by Dave Taht
It would be nice to have a set of official statements from vendors
regarding their bufferbloat solutions vis a vis the various RFCs
(notably rfc8290) before the ietf blog entry goes out.
Years ago, I was trying to get ISOC's Deploy360 folks to help with doing
this.
Post by Dave Taht
My interchange with the tomahawk 3 guy was fruitless. Apple's kind
of exited the router market. Cable manufacturers are obligated to
implement docsis-pie in 3.1. Microsoft has made a big push recently to
modernize their TCPs. A bunch of router makers did embrace fq_codel
(calling it various things like streamboost, adaptive qos, etc), and a
few calling it what it is notably evenroute, and ubnt in their
edgerouter products, sophos, all of lede/openwrt/asuswrt/etc... but
I'm still unaware of any implementations on ONTs, CMTSes, DSLAMs, DSL
firmwares, MOCA...
What I kind of envision is everybody here annoying their favorite
forums for their favorite product and asking what of the outputs of
the ietf aqm working group they planned to implement.
Yeah... So I went through IETF types and through sales channels at one
point, and despite knowing that companies in questions were working on
solutions, nobody knew anything.

--
] Never tell me the odds! | ipv6 mesh networks [
] Michael Richardson, Sandelman Software Works | network architect [
] ***@sandelman.ca http://www.sandelman.ca/ | ruby on rails [
Loading...