Date: Sat, 9 Aug 2014 22:57:29 +0200 From: Michael Tuexen <Michael.Tuexen@lurchi.franken.de> To: hiren panchasara <hiren.panchasara@gmail.com> Cc: "freebsd-net@freebsd.org" <freebsd-net@freebsd.org>, John-Mark Gurney <jmg@funkthat.com>, Niu Zhixiong <kaiaixi@gmail.com>, Bill Yuan <bycn82@gmail.com> Subject: Re: A problem on TCP in High RTT Environment. Message-ID: <7D6601F0-268C-4615-8243-9020499D68B0@lurchi.franken.de> In-Reply-To: <CALCpEUHJ5X%2BhFShgeM4CBcyVaJaQf12WVPWj4b=-qAXSC_48SQ@mail.gmail.com> References: <CAOENNMA_CiBDJc0kchzUbTcf_JBwTJPF=PdBAUB6FPo-KzYkeQ@mail.gmail.com> <20140809184232.GF83475@funkthat.com> <8AE1AC56-D52F-4F13-AAA3-BB96042B37DD@lurchi.franken.de> <CALCpEUHJ5X%2BhFShgeM4CBcyVaJaQf12WVPWj4b=-qAXSC_48SQ@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 09 Aug 2014, at 22:22, hiren panchasara <hiren.panchasara@gmail.com> = wrote: > On Sat, Aug 9, 2014 at 12:51 PM, Michael Tuexen > <Michael.Tuexen@lurchi.franken.de> wrote: >>=20 >> On 09 Aug 2014, at 20:42, John-Mark Gurney <jmg@funkthat.com> wrote: >>=20 >>> Niu Zhixiong wrote this message on Fri, Aug 08, 2014 at 20:34 +0800: >>>> Dear all, >>>>=20 >>>> Last month, I send problems related to FTP/TCP in a high RTT = environment. >>>> After that, I setup a simulation environment(Dummynet) to test TCP = and SCTP >>>> in high delay environment. After finishing the test, I can see TCP = is >>>> always slower than SCTP. But, I think it is not possible. (Plz see = the >>>> figure in the attachment). When the delay is 200ms(means = RTT=3D400ms). >>>> Besides, the TCP is extremely slow. >>>>=20 >>>> ALL BW=3D20Mbps, DELAY=3D 0 ~ 200MS, Packet LOSS =3D 0 (by = dummynet) >>>>=20 >>>> This is my parameters: >>>> FreeBSD vfreetest0 10.0-RELEASE FreeBSD 10.0-RELEASE #0: Thu Aug 7 >>>> 11:04:15 HKT 2014 >>>>=20 >>>> sysctl net.inet.tcp >>>=20 >>> [...] >>>=20 >>>> net.inet.tcp.recvbuf_auto: 0 >>>=20 >>> [...] >>>=20 >>>> net.inet.tcp.sendbuf_auto: 0 >>>=20 >>> Try enabling this... This should allow the buffer to grow large = enough >>> to deal w/ the higher latency... >>>=20 >>> Also, make sure your program isn't setting the recv buffer size as = that >>> will disable the auto growing... >> I think the program sets the buffer to 2MB, which it also does for = SCTP. >> So having both statically at the same size makes sense for the = comparison. >> I remember that there was a bug in the combination of LRO and delayed = ACK, >> which was fixed, but I don't remember it was fixed before 10.0... >=20 > If you are thinking of r256920, I believe it did make it into 10.0R. Yepp, that is what I was thinking of... Best regards Michael >=20 > cheers, > Hiren >=20
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?7D6601F0-268C-4615-8243-9020499D68B0>