From mboxrd@z Thu Jan 1 00:00:00 1970 MIME-Version: 1.0 In-Reply-To: References: Date: Wed, 17 Aug 2011 22:48:26 -0700 Message-ID: From: John Floren To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net> Content-Type: multipart/alternative; boundary=0016e652e24e5ac49004aac129e3 Subject: Re: [9fans] Help with two small shared file servers Topicbox-Message-UUID: 143152ce-ead7-11e9-9d60-3106f5b1d025 --0016e652e24e5ac49004aac129e3 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable On Aug 17, 2011 10:40 PM, "erik quanstrom" wrote: > > slower than transferring via HTTP. When I tested 9P vs. HTTP over > > connections with 25 ms latency (50ms RTT), I saw a 4x slowdown versus > > HTTP. Even at 15 ms RTT, transfers took twice as long. > > did you do testing at regular lan latencies? what i see on my networks > is usually <=3D 30=B5s. > > - erik > I induced latencies similar to those found on the Internet. My tests at LAN speeds (500us in my sub-optimal setup) had 9P at essentially the same transfer rate as HTTP. My work was motivated by the observation that transferring files to and from sources is just deathly slow, and that while fcp does a good job helping that, it is just a patch covering a deeper issue. --0016e652e24e5ac49004aac129e3 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable


On Aug 17, 2011 10:40 PM, "erik quanstrom" <quanstro@quanstro.net> wrote:
> > slower than transferring via HTTP. When I tested 9P vs. HTTP over=
> > connections with 25 ms latency (50ms RTT), I saw a 4x slowdown ve= rsus
> > HTTP. Even at 15 ms RTT, transfers took twice as long.
>
> did you do testing at regular lan latencies? =A0what i see on my netwo= rks
> is usually <=3D 30=B5s.
>
> - erik
>

I induced latencies similar to those found on the Internet. My tests at = LAN speeds (500us in my sub-optimal setup) had 9P at essentially the same t= ransfer rate as HTTP. My work was motivated by the observation that transfe= rring files to and from sources is just deathly slow, and that while fcp do= es a good job helping that, it is just a patch covering a deeper issue.

--0016e652e24e5ac49004aac129e3--