Message321726
| Author |
ammar2 |
| Recipients |
ammar2, giampaolo.rodola, jkloth, paul.moore, steve.dower, tim.golden, vstinner, zach.ware |
| Date |
2018-07-16.09:36:43 |
| SpamBayes Score |
-1.0 |
| Marked as misclassified |
Yes |
| Message-id |
<1531733803.29.0.56676864532.issue34060@psf.upfronthosting.co.za> |
| In-reply-to |
|
| Content |
I don't think taking instantaneous values instead of averaging will work out too well. For reference I've attached a screenshot. It has sampled values at every second on an unloaded computer and then with running prime95 for cpu stress testing. The load tends to peak and fall.
>Is it exactly the same thing on Unix (load average)?
Indeed it is: https://en.wikipedia.org/wiki/Load_(computing)#Unix-style_load_calculation
"An idle computer has a load number of 0 (the idle process isn't counted). Each process using or waiting for CPU (the ready queue or run queue) increments the load number by 1."
From what I can tell, the number of processors are dealt with the same way as on Linux, that is, a single core processor is overloaded when the load is >1 and a quad core processor is overloaded when the load is >4 |
|