From: Jim M. <jim...@ex...> - 2004-08-05 15:52:14
|
Jim Mozley wrote: > So I need to script a wrapper to set the correct -b option given a -l > value so I allow for the overhead, assuming I want a constant wire rate. From a previous post http://archive.ncsa.uiuc.edu/lists/iperf-users/jan03/msg00023.html I see that "There is no way to really before hand compute the amount of headers a piece of information will get covered with." So setting -b for different lengths of buffer ( -l option ) is a trial and error process and not something that can be calculated :-(. This would explain why my calculations and observable results are different. I'm guessing different operating systems and network card combinations for the Iperf testers will behave differently in this regard. Jim Mozley |