Re: Issues with a Large Fat pipe Network simulation

From: Luigi Rizzo (rizzo_at_icir.org)
Date: 06/20/05

  • Next message: Pieter de Boer: "Re: Issues with a Large Fat pipe Network simulation" 
    Date: Mon, 20 Jun 2005 13:50:44 -0700
To: Pieter de Boer <pieter@thedarkside.nl>

    On Mon, Jun 20, 2005 at 10:11:27PM +0200, Pieter de Boer wrote:
    > Hello there,
    ...
    > When testing without any extra delay on 'network' and send/recvspaces of
    > 65535 bytes, we can sustain around 800mbit/s. The interrupts on
    > 'network' may be the limiting factor here. However, when we set the
    > send/recv space to 65535*2, we can only sustain around 200-300mbit/s. It

    have you checked that these two
            kern.ipc.maxsockbuf: 262144
            kern.ipc.sockbuf_waste_factor: 8

    aren't responsible for the trouble ?
    I won't enter into details because i am not sure but try upping
    maxsockbuf might help. Or not. in any case it would be nice to try
    and let us know

            cheers
            luigi

    > seems the speed isn't as 'stable' either (peaks of more than 300mbit/s,
    > sometimes up to 500mbit/s). We also used read/write sizes of 128KB using
    > the -l option on iperf, but this didn't seem to have any noticeable
    > effect.
    >
    > When adding extra latency on 'network' and adjusting the
    > send/recv-spaces to correct for the greater bandwidth*delay product, we
    > weren't able to sustain rates much higher than 200mbit/s either.
    >
    >
    > Can anyone shed some light on what we're seeing here?
    >
    > --
    > With regards,
    > Pieter de Boer
    >
    > _______________________________________________
    > freebsd-net@freebsd.org mailing list
    > http://lists.freebsd.org/mailman/listinfo/freebsd-net
    > To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org"
    _______________________________________________
    freebsd-net@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-net
    To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org"

  • Next message: Pieter de Boer: "Re: Issues with a Large Fat pipe Network simulation"

    Relevant Pages

    • Re: [PATCH] [Request for inclusion] Filesystem in Userspace
      ... > of writable pages assigned to FUSE filesystems. ... Limiting this should ... > solve the deadlock problem. ... However, when receiving the ACK network packets from the server, ...
      (Linux-Kernel)
    • Re: Blaster Worm.. Please help
      ... Select the advanced tab -> Enable "Protect my computer and network by ... limiting or preventing access to this computer from the internet" ...
      (microsoft.public.security.virus)
    • Issues with a Large Fat pipe Network simulation
      ... to simulate a network with a high bandwidth*delay product. ... We're testing using iperf ... When testing without any extra delay on 'network' and send/recvspaces of ... we can only sustain around 200-300mbit/s. ...
      (freebsd-net)
    • Re: how tunnel X11 to a slow system?
      ... I also did try a faster cipher, ... The reason why I suspect that is not related to network speed is that as ... would think that 180Mhz would be enough to handle the SSH encryption but ... > be true, but your network guys may be limiting your port to 10 MB, ...
      (comp.security.ssh)
    • Re: how tunnel X11 to a slow system?
      ... Before you try some confusing scenario, try running SSH on system B ... This will compress the data before it is sent ... Also realize that X apps can be very network intensive and the speed ... be true, but your network guys may be limiting your port to 10 MB, ...
      (comp.security.ssh)