I can't answer your question regarding the innards of NISTNET, but I can
address the delay question (having recently looked at some field data).

 

During an experiment our team conducted recently, one-way delays on the
public internet averaged around 40 milliseconds, with peak measured delays
being several SECONDS long (the raw data could be accurately modeled using a
shifted lognormal density - furthermore, there was a slowly decaying
autocorrelation in the traffic).   After messing around with NISTNET's
parameters, we could duplicate the delay density, but not the
autocorrelation in the delays (NISTNET appears to use some sort of Markov
process for generating the autocorrelation - hence an exponentially decaying
autocorrelation whereas field data decayed more slowly).

 

So, depending on the congestion in the network, I feel a 400 ms delay may be
perfectly reasonable.

 

Warmest regards,

 

Bo

 

 

  _____  

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
On Behalf Of Xiaochuan Shen
Sent: Thursday, November 08, 2007 7:44 AM
To: nistnet@antd.nist.gov
Cc: 'Adrian Wonfor'
Subject: [nistnet] how much buffer NISTNet use for incoming traffic

 

Hello group,

 

I am just wondering if anyone got any idea how much buffer NistNet uses (or
in what way it determines how much buffer should be used) when bandwidth
limitation is applied.

 

 

This is what I implemented:

 

 

server ----------------(100mb link) ---------------- NistNet box
------------------(100mb link) --------------------- Client

 

 

I am transferring bulk TCP traffic from server to client (a huge file using
FTP transfer). 

 

If I apply:

Cnistnet server client -bandwidth 187500

 

Then I observed the "queue length" is around 50 packets. 

 

If all of these packets are full packet of size 1500Byte, a queue length of
50 packets will introduce 50 * 1500 / 187500 = 400 milli-seconds delay in my
configuration. Is this value realistic in a real network? If not, why
nistnet use such a length of buffer?

I can understand nistnet has to buffer a number of packets before it can
process the coming traffic but is this queue length configurable in nistnet?

 

Thanks for anyone looking and answering!

 

Regards,

Xiaochuan

_______________________________________________
nistnet mailing list
nistnet@antd.nist.gov
http://www-x.antd.nist.gov/mailman/listinfo/nistnet

Reply via email to