can someone explain load average in flower?

Question

I use flower to monitor my rabbitmq queues, I am not able to understand how load average is calculated, if someone can explain then that would be of great help.
I've a quad core processor .
Thank You.

Looking through the celery & flower source, it ultimately comes from here https://github.com/celery/celery/blob/c122150887ae3633ff0164a5670d23bd093354e4/celery/utils/sysinfo.py#L13-L14 — Stephen Fuhry, Jan 20 '17 at 13:25

score 14 · Answer 1 · answered Mar 06 '18 at 16:34

14

I understand that this question has been asked for awhile and that it's probably figured out but for those who are new here is what I found:

As Stephen pointed out in the comments load average is defined as follows :

def _load_average():
    return tuple(ceil(l * 1e2) / 1e2 for l in os.getloadavg())

The source file is here.

And os.getloadavg() as documented here:

system run queue averaged over the last 1, 5, and 15 minutes

Therefore the three numbers in the load average in the flower dashboard is just the system's queue load average multiplied by 100 (percentage, I am guessing) over the last 1, 5, and 15 minutes, respectively.

answered Mar 06 '18 at 16:34

Allen

171
1
6

1

It multiplied by 100 only for the ceiling and right after it divides by 100. That means the the value we see it the value itself, not multiplied by 100 – Igal Drayerman Apr 22 '21 at 13:20
So queue load average is the average number of tasks in the queue ? – Florent Dec 08 '22 at 09:54

score 1 · Answer 2 · answered Oct 05 '21 at 12:27

It denoted the load for specific worker/worker system in three different timelines. Consider this is the load average you are seeing in screen, 0.44, 0.28, 0.25, then it explains this way.

last minute (0.44)
last five minutes (0.28)
the last fifteen (0.25).

This article will give complete clarity on this topic.

can someone explain load average in flower?

2 Answers2