Difference in iperf output between client and server

iperf

Been playing with iperf for the last 4 hours and I'm unable to understand why the results are different in client and server.

Consider the following iperf output on client :

iperf -c 91.121.112.163 -p 2222
------------------------------------------------------------
Client connecting to 91.121.112.163, TCP port 2222
TCP window size: 1.06 MByte (default)
------------------------------------------------------------
[  3] local 10.0.0.160 port 36726 connected with 91.121.112.163 port 2222
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0-11.4 sec  1.88 MBytes  1.38 Mbits/sec

And here is the corresponding output on server :

iperf -V -s -p 2222
------------------------------------------------------------
Server listening on TCP port 2222
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[  4] local ::ffff:91.121.112.163 port 2222 connected with ::ffff:86.209.154.156 port 36726
[ ID] Interval       Transfer     Bandwidth
[  4]  0.0-19.1 sec  1.88 MBytes   823 Kbits/sec

So, the client is uploading at 1.38Mbits/sec but the server is receiving at 823kbits/sec ?

Do someone have an explanation about these confusing results ?

Best Answer

From the client: (emphasis mine)

[ ID] Interval       Transfer     Bandwidth
[  3]  0.0-**11.4** sec  1.88 MBytes  1.38 Mbits/sec

From the server:

[ ID] Interval       Transfer     Bandwidth
[  4]  0.0-**19.1** sec  1.88 MBytes   823 Kbits/sec

For some reason, the server thinks the test is taking ~8 seconds longer than the client does. That would be enough to drop the overall bandwidth (from the server's perspective) to 823 Kbps.

Related Solutions

Windows – Different versions of iperf for windows give totally different results

The later version of iperf you have is using a different threading library (win32 threads vs. pthread). That alone could explain the difference in performance.

There could be some other code difference between 1.x and 2.x of iperf. It could be simple like the default options changed from 1.x to 2.x (such as TCP_WINDOW_SIZE, ...).

Make sure both are using TCP (or udp), for example by re-running with the --udp option.

You might try the iperf-users mailing list archives, or even posting your question there.

How does iperf calculate throughput and jitter

Is it possible to gather the same results by sending regular data, as in not testing data?

I don't believe so, at least not using iperf. If you can capture the real data with pcap/tcpdump/wireshark, then you should be able to use wireshark to determine the jitter.

I suspect iperf calculates jitter using a formula like is defined in RFC 3550 for the RTP protocol. You would probably need to dig into the source to know for certain.

  If Si is the RTP timestamp from packet i, and Ri is the time of
  arrival in RTP timestamp units for packet i, then for two packets
  i and j, D may be expressed as

     D(i,j) = (Rj - Ri) - (Sj - Si) = (Rj - Sj) - (Ri - Si)

  The interarrival jitter SHOULD be calculated continuously as each
  data packet i is received from source SSRC_n, using this
  difference D for that packet and the previous packet i-1 in order
  of arrival (not necessarily in sequence), according to the formula

     J(i) = J(i-1) + (|D(i-1,i)| - J(i-1))/16

  Whenever a reception report is issued, the current value of J is
  sampled.

  The jitter calculation MUST conform to the formula specified here
  in order to allow profile-independent monitors to make valid
  interpretations of reports coming from different implementations.

Related Topic