[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ns] using "ndatapack" as throughput measurement of TCP agent

To: Jake Whitehill <whitehil@CS.Stanford.EDU>
Subject: Re: [ns] using "ndatapack" as throughput measurement of TCP agent
From: Lloyd Wood <l.wood@eim.surrey.ac.uk>
Date: Mon, 29 Jan 2001 11:23:59 +0000 (GMT)
cc: ns-users@ISI.EDU
In-Reply-To: <Pine.GSO.4.21.0101281805500.7241-100000@Xenon.Stanford.EDU>
Organization: speaking for none
Reply-To: L.Wood@eim.surrey.ac.uk
Sender: owner-ns-users@ISI.EDU

On Sun, 28 Jan 2001, Jake Whitehill wrote:

> If I wanted to determine how many bytes were received at the destination
> of a  TCP connection, can I just examine "ndatapack" in the sender?

No. Please see my previous discussion of adding a recvBytes method in
tcp-sink.cc(), diffs posted to this mailing list, etc.

> I'm
> pretty sure that this will not yield exact results because this includes
> packets that are retransmitted, I believe. But will it yield close
> results?

No.

L.

<L.Wood@surrey.ac.uk>PGP<http://www.ee.surrey.ac.uk/Personal/L.Wood/>

References:
- [ns] using "ndatapack" as throughput measurement of TCP agent
  - From: Jake Whitehill <whitehil@CS.Stanford.EDU>

Prev by Date: [ns] PATCH: nse on Linux routing Bug/Problem
Next by Date: [ns] Random Exponential Marking
Prev by thread: [ns] using "ndatapack" as throughput measurement of TCP agent
Next by thread: [ns] Scheduler backwards problem
Index(es):
- Date
- Thread