Re: [Bloat] hystart++
The puzzling thing is turning off pacing in these implementations... On Tue, May 30, 2023, 9:31 AM Joerg Deutschmann via Bloat < bloat@lists.bufferbloat.net> wrote: > And of course some QUIC implementations have HyStart++, e.g., > https://blog.cloudflare.com/cubic-and-hystart-support-in-quiche/ > > > On 30.05.23 17:27, Joerg Deutschmann via Bloat wrote: > > Hi Dave, > > > > some brief information regarding HyStart++ on Windows: > > > https://techcommunity.microsoft.com/t5/networking-blog/algorithmic-improvements-boost-tcp-performance-on-the-internet/ba-p/2347061 > > > https://datatracker.ietf.org/meeting/113/materials/slides-113-tcpm-draft-ietf-tcpm-hystartplusplus-04-00 > > > > AFAIK, there is no Linux implementation for HyStart++, only HyStart: > > > https://www.kernel.org/doc/html/latest/networking/snmp_counter.html#tcp-hybrid-slow-start > > (but I would be interested if there was a HyStart++ implementation for > > Linux; also there are several papers about HyStart but not HyStart++) > > > > And HyStart++ should be available in FreeBSD: > > https://reviews.freebsd.org/D33035 > > > https://cgit.freebsd.org/src/commit/?id=653cf466f08584c0fd87bc336cdc1db08f0d4434 > > > > Best regards, > > Joerg > > > > > > On 28.05.23 14:46, Dave Taht via Bloat wrote: > >> Does a linux implementation of this exist? It looks promising... > >> > >> https://datatracker.ietf.org/doc/html/rfc9406 > >> > > > > ___ > > Bloat mailing list > > Bloat@lists.bufferbloat.net > > https://lists.bufferbloat.net/listinfo/bloat > ___ > Bloat mailing list > Bloat@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/bloat > ___ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat
Re: [Bloat] hystart++
And of course some QUIC implementations have HyStart++, e.g., https://blog.cloudflare.com/cubic-and-hystart-support-in-quiche/ On 30.05.23 17:27, Joerg Deutschmann via Bloat wrote: Hi Dave, some brief information regarding HyStart++ on Windows: https://techcommunity.microsoft.com/t5/networking-blog/algorithmic-improvements-boost-tcp-performance-on-the-internet/ba-p/2347061 https://datatracker.ietf.org/meeting/113/materials/slides-113-tcpm-draft-ietf-tcpm-hystartplusplus-04-00 AFAIK, there is no Linux implementation for HyStart++, only HyStart: https://www.kernel.org/doc/html/latest/networking/snmp_counter.html#tcp-hybrid-slow-start (but I would be interested if there was a HyStart++ implementation for Linux; also there are several papers about HyStart but not HyStart++) And HyStart++ should be available in FreeBSD: https://reviews.freebsd.org/D33035 https://cgit.freebsd.org/src/commit/?id=653cf466f08584c0fd87bc336cdc1db08f0d4434 Best regards, Joerg On 28.05.23 14:46, Dave Taht via Bloat wrote: Does a linux implementation of this exist? It looks promising... https://datatracker.ietf.org/doc/html/rfc9406 ___ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat smime.p7s Description: S/MIME Cryptographic Signature ___ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat
Re: [Bloat] hystart++
Hi Dave, some brief information regarding HyStart++ on Windows: https://techcommunity.microsoft.com/t5/networking-blog/algorithmic-improvements-boost-tcp-performance-on-the-internet/ba-p/2347061 https://datatracker.ietf.org/meeting/113/materials/slides-113-tcpm-draft-ietf-tcpm-hystartplusplus-04-00 AFAIK, there is no Linux implementation for HyStart++, only HyStart: https://www.kernel.org/doc/html/latest/networking/snmp_counter.html#tcp-hybrid-slow-start (but I would be interested if there was a HyStart++ implementation for Linux; also there are several papers about HyStart but not HyStart++) And HyStart++ should be available in FreeBSD: https://reviews.freebsd.org/D33035 https://cgit.freebsd.org/src/commit/?id=653cf466f08584c0fd87bc336cdc1db08f0d4434 Best regards, Joerg On 28.05.23 14:46, Dave Taht via Bloat wrote: Does a linux implementation of this exist? It looks promising... https://datatracker.ietf.org/doc/html/rfc9406 smime.p7s Description: S/MIME Cryptographic Signature ___ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat
Re: [Bloat] hystart++
btw, this might make you happy. Adding about 7k devices a week, some signs it is beginning to go exponential. Got any ideas to help make it go exponential? :) http://stats.libreqos.io:3000/ On Mon, May 29, 2023 at 7:38 PM Stephen Hemminger wrote: > > On Sun, 28 May 2023 06:46:48 -0600 > Dave Taht via Bloat wrote: > > > Does a linux implementation of this exist? It looks promising... > > > > https://datatracker.ietf.org/doc/html/rfc9406 > > > > Not that I know of. Contact the authors and they may be able to help. > Microsoft is more open source friendly now, but obviously all the research is > on Windows implementation. -- Podcast: https://www.linkedin.com/feed/update/urn:li:activity:7058793910227111937/ Dave Täht CSO, LibreQos ___ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat
Re: [Bloat] hystart++
On Sun, 28 May 2023 06:46:48 -0600 Dave Taht via Bloat wrote: > Does a linux implementation of this exist? It looks promising... > > https://datatracker.ietf.org/doc/html/rfc9406 > Not that I know of. Contact the authors and they may be able to help. Microsoft is more open source friendly now, but obviously all the research is on Windows implementation. ___ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat