How is iPerf used to measure delay?

5 pts.
Tags:
iPerf
Networking
How does iPerf measure end-to-end delay? If delay is between h1 and si1 is 2ms and s1 and h2 is 3ms today delay in output should be 5ms?

Software/Hardware used:
linux iperf python

Answer Wiki

Thanks. We'll let you know when a new response is added.

There are three types of “delay” measurements available.  This is from an  iperf 2.0.10  perspective. 

  1. UDP jitter (delay variance) – requires one clock
  2. TCP Round Trip Time (RTT) – requires one clock
  3. UDP end/end delay – requires two clocks to be synchronized
The latter two require the -e option to get the stats.

The iperf client inserts a microsecond timestamp in every UDP packet.   When the client and server clocks are synchronized the UDP end/end delay is valid.   I use an oven control oscillator as a PTP grandmaster and then use PTP to synchronize the clocks.

For TCP, iperf samples the kernels view of the RTT. For UDP jitter, the variance is calculated by the client using the clock from the server.


(Note that iperf 2.10.10 will use realtime clocks, which are monotonic, when possible.)

Bob

Discuss This Question:  

 
There was an error processing your information. Please try again later.
Thanks. We'll let you know when a new response is added.
Send me notifications when members answer or reply to this question.

Forgot Password

No problem! Submit your e-mail address below. We'll send you an e-mail containing your password.

Your password has been sent to:

To follow this tag...

There was an error processing your information. Please try again later.

Thanks! We'll email you when relevant content is added and updated.

Following

Share this item with your network: