Shopping cart    |      
Results 1 to 4 of 4

Thread: throughput measurements

  1. #1
    Join Date
    Jun 2011
    Posts
    3

    Default throughput measurements

    I am using Iperf to send data through a wireless network and i am using wirelessmon to log the throughput. For some odd reason when wirelessmon is open the throughput on iperf on an excellent signal bounces up and down crazily. On the other hand, if i close WirelessMon the jperf throughput shows steadily as expected. Any reasons as to why when wirelessmon is open it casues the throughput to bounce.

    Do you think its because wirelessmon may duty cycle with the radio signal???

  2. #2
    Join Date
    Jan 2003
    Location
    Sydney Australia
    Posts
    4,183

    Default

    Probably poor design on the wireless card. In that it can't scan the available access points and send data at the same time.

    Try reducing the sample frequency in WirelessMon.

  3. #3
    Join Date
    Jun 2011
    Posts
    3

    Default

    thks for the prompt reply....that makes sense, i have been using a g-card and a g-usb dongle and compared results to the same negative effects also i have ran tests using a b-card and everything runs very smoothly making your assumption valid....i have reduced the sampling frequency and am now obtaining correct throughput measurements for both adapters....guess it seems as though this will be the only way to get correct data even though the process will take much longer......
    any suggestions to speed up the process???

    possible solutions i am currently thinking of:
    1.) replace card or dongle
    2.) obtain different driver for card or dongle
    3.) instead of sending TCP packets, send UDP packets
    4.) figure out way to reduce load on CPU
    5.) open ears to any suggestions you may have

    thanks for your help....

  4. #4
    Join Date
    Jan 2003
    Location
    Sydney Australia
    Posts
    4,183

    Default

    I don't understand the problem.
    If you have already taken your measurements, and found the result to be correct, then why are you now looking for a quicker way.

    Changing the sample rate for, for example, 1 second to 10 seconds, should only add 10 secs to your measurement time.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •