2024-10-30T11:19:27.744Z

<Jose J Palacios-Perez> Hi <@U04JNK33PE0>, some quick questions please: for the target of max number of seastar reactor threads on the system: 2 sockets, 28 physical cores per socket, 56 per socket including hyperthreading: 112 total CPU reported by lscpu.
• Reserve 8 CPUs for the client FIO, so we have two cases:
• physical: that leaves 24 physical cores per socket for both seastar reactors and alien threads,
• hyperthreading: 52 CPU per socket
Here is an example of CPU distribution scenarios: what are your thoughts please? Thanks in advance: https://files.slack.com/files-pri/T1HG3J90S-F07UH4G82JV/download/screenshot_2024-10-30_at_11.17.32.png

2024-10-30T14:33:37.747Z

<Matan Breizman> This looks better, thanks for updating!

In the multiple OSD case (1, 3 or 8) when specifying Seastar Cores, perhaps it will be more understandable to write: `crimson cpus` * `OSD num` / `alien cpus`
For example, for 24 physical and 3 OSD: `5*3 / 9`

Why shouldn't we use both sockets? It seems that the tests are limited to a single socket? If not, should we instead share the total resources available (by both sockets)?