Re: tcptrace Slow Start Duration

From: Hans-Werner Braun (hwb@nlanr.net)
Date: 03/26/03

  • Next message: Manikantan Ramadas: "CVS repository access to tcptrace"

    Message-ID: <20030326162447.12233@nlanr.net>
    Date: Wed, 26 Mar 2003 16:24:47 -0800
    From: Hans-Werner Braun <hwb@nlanr.net>
    Subject: Re: tcptrace Slow Start Duration
    
    

    >From my experience in wireless networks the case that most
    >frequently happens is "window-limited" and that is also the only
    >one I would be interested in. Losses tend not to occur as buffering
    >is used at the border between wired and wireless, the other 2 also
    >would not be relevant to answer the question "for how long the
    >bandwidth is under-utilised".

    I see plenty loss in wireless networks, depending on what kind of
    radio you use and what your fade margin and environmental conditions
    are. E.g., 802.11b retransmits (except multicast/broadcast). various
    networks use FEC and fall over at 10^-6 or so. Some do not use
    retransmissions or FEC, and clobbered packets don't get through.

    We moved from networks in the 1980s that had plenty of messed up
    packets to high reliability on fiber networks in the 1990s. Now we are
    more back to less reliability on wireless substrates. Can't be as
    pointy as a laser on wireless RF networks, and interference can happen
    easily, especially in license-exempt environments, which can mess up
    packets and cause losses easily due to interference.
    ----------------------------------------------------------------------------
    To unsubscribe, send a message with body containing "unsubscribe tcptrace" to
    majordomo@tcptrace.org.



    This archive was generated by hypermail 2b30 : 03/27/03 EST