Date: Fri, 12 Oct 2001 11:56:32 -0400 From: Mike Tancsa <mike@sentex.net> To: rizzo@aciri.org Cc: freebsd-net@FreeBSD.ORG Subject: Re: strange results with increased net.inet.ip.intr_queue_maxlen Message-ID: <5.1.0.14.0.20011012114959.0717a2a0@marble.sentex.ca> In-Reply-To: <200110120130.f9C1Ubw15419@iguana.aciri.org> References: <200110120116.f9C1GEv18196@arch20m.dellroad.org>
next in thread | previous in thread | raw e-mail | index | archive | help
At 06:30 PM 10/11/01 -0700, Luigi Rizzo wrote:
> > > from pinging the other side of the OC-3 or ethernet connection and
> > > measuring the response time, how can I see how much latency is added by
> > > increasing these buffers ?
>
>of course the latency increase depends on how full are the buffers,
>and the worst case is easier to determine by back-of-the-envelope
>calculations:
>
> queue_slots * max_pkt_size / bottleneck_link_speed
>
>e.g. if you have 100 slots and an MSS of 1500 and a 10Mbit
>bottleneck you are adding (100*1500*8 / 10000000) = 120ms latency.
Interesting! In my case, one of the machines is oc-3 and ther other,
FastE. I guess now it becomes a choice of what is the best choice for my
situation which I guess is not so easy to figure out. Is it better to drop
packets when the queue is full and let the various applications behind me
figure it out, or is it better to add some latency at the network layer so
the apps dont have to deal with it.
Forgive me if my questions are simplistic, I am just trying to get as much
info to make a more informed decision as how to best configure my network
given the resources I have.
---Mike
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5.1.0.14.0.20011012114959.0717a2a0>