Also I think, if the total rate (including your TCP or UDP packets plus
all lower layer headers which are quite substantial in wireless MAC
protocols is greater than the available data rate (rate at which packets
are pushed out into the channel), no matter how large your IFQ buffer
is, sooner or later there will be lots of IFQ drops and hence a low
throughput. Let me know what you feel. I have experimented with some UDP
traffic in 802.11b and if the rate is low enough, there will not be any
IFQ drops for IFQ buffers as small as 10-20 bytes after the simulation
has stabilized. When the rate is large buffers of size 10 or 1000 did
not matter. Let me know what you feel.

Regards
Arun


-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of mix si
Sent: Thursday, December 29, 2005 5:19 PM
To: Dario Borriello
Cc: [email protected]
Subject: Re: [ns] how to avoid DROP IFQ ARP


Hello Dario,
  
  try this link:
  
  http://mailman.isi.edu/pipermail/ns-users/2002-March/020912.html
  
  i think you will find this page useful!
  
  I wish you a happy new year!!
  regards,
  
  Michael Sidiropoulos

Dario Borriello <[EMAIL PROTECTED]> wrote:  
Hi,

i'm making a simulation on wireless node and i have noticed that the 
throughput is very limitated due to DROP IFQ ARP.
i have read that If when one packet is waiting for ARP reply, another
packet 
which has the same destination address as the first arrives in the
queue,the 
second packet will overwrite the first and send another ARP request at
once.
Is there a way to avoid this problem?
Or it is impossible to increase the throughput?

Thanks

_________________________________________________________________
250MB per la tua casella di posta http://www.msn.it/hotmail/minisite_10 
Trova immediatamente qualsiasi tipo di file.




                
---------------------------------
Yahoo! Shopping
 Find Great Deals on Holiday Gifts at Yahoo! Shopping 



DISCLAIMER:
This email (including any attachments) is intended for the sole use of the 
intended recipient/s and may contain material that is CONFIDENTIAL AND PRIVATE 
COMPANY INFORMATION. Any review or reliance by others or copying or 
distribution or forwarding of any or all of the contents in this message is 
STRICTLY PROHIBITED. If you are not the intended recipient, please contact the 
sender by email and delete all copies; your cooperation in this regard is 
appreciated.

Reply via email to