Timeline for Does it ever make sense to use more concurrent processes than processor cores?
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Sep 24, 2020 at 7:12 | history | edited | Glorfindel | CC BY-SA 4.0 | Fixed typos & punctuation |
| S Sep 24, 2020 at 7:12 | history | suggested | sdore | CC BY-SA 4.0 | Fixed typos & punctuation |
| Sep 23, 2020 at 10:26 | review | Suggested edits | |||
| S Sep 24, 2020 at 7:12 | |||||
| Sep 1, 2020 at 17:36 | comment | added | Wayne Conrad | +1 for profiling. When you plot throughput vs workers, there's usually a linear part of the curve, and then a part where it starts to flatten out or even decreases. Just before that knee where it flattens out is often where you want to be. And certainly not where throughput decreases. Without profiling, you don't know if the added workers are helping, doing no good, or actually hurting. | |
| Sep 1, 2020 at 16:08 | review | First posts | |||
| Sep 2, 2020 at 5:40 | |||||
| Sep 1, 2020 at 16:03 | history | answered | Philippe Verdy | CC BY-SA 4.0 |