From: Bu, Tian (Tian) (tbu@research.bell-labs.com)
Date: 06/02/05
Message-ID: <3BD3BB8D47F9DE4984C319EA325A006D444AEA@nj9620exch003u.mh.lucent.com> From: "Bu, Tian (Tian)" <tbu@research.bell-labs.com> Subject: RE: tcptrace-bugs rtt estimate Date: Thu, 2 Jun 2005 11:13:39 -0400
Thanks for the answer. This is very helpful.
-Tian
> -----Original Message-----
> From: Shawn Ostermann [mailto:sdo@picard.cs.ohiou.edu]
> Sent: Thursday, June 02, 2005 10:48 AM
> To: Manikantan Ramadas
> Cc: Bu, Tian (Tian); 'tcptrace-bugs@tcptrace.org'
> Subject: Re: tcptrace-bugs rtt estimate
>
>
> It calculates the time between when a segment was sent and when the ACK
> for it was received. So, technically, it's the RTT between the
> measurement host and the data receiver. As Mani pointed out, the
> accuracy of that will depend on the accuracy of the clock that the
> measurement host is using and that pcap is putting into the packet
> headers for the captured packets.
>
> --sdo
>
>
>
> Manikantan Ramadas <mramadas@masaka.cs.ohiou.edu> wrote:
>
> > tcptrace's rtt estimates are based on the pcap dumpfile, and therefore
> > directly depend on where the dumpfile was captured, i.e., where the
> > measurement host is.
> >
> > - Mani.
> >
> > On Wed, Jun 01, 2005 at 09:58:12AM -0400, Bu, Tian (Tian) wrote:
> > > Hello there,
> > >
> > >
> > >
> > > As tcptrace estimates the rtt, is this estimate for end-to-end rtt or
> the rtt between the measurement point and the destination?
> > >
> > >
> > >
> > > Thanks!
> > >
> > > -Tian
> > >
> >
> > --
> > "'Beauty is truth, truth beauty,'--that is all
> > Ye know on earth, and all ye need to know." - John Keats
> > ____________________________________________________________________
> >
> > * Manikantan Ramadas * IRG, OU * http://irg.cs.ohiou.edu/~mramadas *
> > ____________________________________________________________________
>
>
This archive was generated by hypermail 2.1.7 : 06/02/05 EDT