Timeline for How to measure actual CPU utilization in Linux for multi core applications?
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 28 at 14:05 | vote | accept | tkw954 | ||
| Apr 24 at 14:43 | history | edited | tkw954 | CC BY-SA 4.0 | added 3298 characters in body |
| Apr 24 at 3:03 | answer | added | user10489 | timeline score: 2 | |
| Apr 24 at 1:09 | comment | added | PersianGulf | I remember htop filter by thread. | |
| Apr 23 at 22:53 | comment | added | Useless | Very likely top is seeing that you have 12 cores available, so when you only start 6 jobs, you naturally get 50% utilization. For scheduling purposes, each "hyperthread" instruction pipeline is treated like a core. | |
| Apr 23 at 19:40 | comment | added | Chester Gillon | To help get an answer, can you edit the question to show the output from top for the different cases? | |
| S Apr 23 at 19:35 | review | First questions | |||
| Apr 23 at 19:40 | |||||
| S Apr 23 at 19:35 | history | asked | tkw954 | CC BY-SA 4.0 |