I am going to close this for now. Two reasons:

1. The primary reason for this report was that I misunderstood how Flent measures latency for its default plot. Currently, that plot shows latency to the -H host. As noted in https://github.com/tohojo/flent/issues/181#issuecomment-530500478, there could be an option to substitute a nearby host for default ping measurements. This would be great because the measured latency would likely have lower variance, and any induced bloat would be a larger percentage of the total.

2. I can no longer reproduce the difference between Flent's reported latency (say, to ping_host) and a separate fping session. My original measurement process might have been faulty - I can no longer say how I got those values.

Thanks.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/tohojo/flent/issues/178#issuecomment-530801400