0

Iperf3 is a great tool and is seen as the gold standard in terms of testing network performance however its enormous scope of functionality makes it difficult to configure for specific use cases.

What are the best settings to configure at both client and server end to most closely test general browser performance and specifically offer a reasonably direct comparison to the major browser test sites such as speedtest.com etc?

Paula Livingstone
  • 1,175
  • 1
  • 12
  • 21

1 Answers1

0

None of the above, IPerf is not a good simulation to browsing performance.

IPerf generates a (more or less) steady state stream of data, while browsing generates more bursty data with different peaks.

Even for speed testing IPerf is less than adequate, read speedtest test description to understand why- they use an adaptive method and not just "as fast as you can".

Rsf
  • 126
  • 1
  • 6
  • iperf3 maintainer here. Thanks for posting the pointer to the Speedtest test description. I would not say that it's better than iperf3, it's just different. The primary (or I should say main intended) use of iperf3 is to measure the best single-stream performance you can get over a wide area network for a given set of parameters. It does that pretty well IMHO. Speedtest.net measures something different, it looks like it's trying its very hardest to saturate a path. That's an equally valid test to do, and from what I've seen, Speedtest.net does a good job of that too. – Bruce A. Mah Oct 11 '17 at 23:23