| kata-irc-bot | <eric.ernst> @archana.m.shinde hey -- I was looking at the static resource mgmt part | 15:34 |
|---|---|---|
| kata-irc-bot | <eric.ernst> https://github.com/kata-containers/kata-containers/commit/8cde54131af6f036e390e7cf8a44c9b8d70d2ff8 | 15:34 |
| kata-irc-bot | <eric.ernst> I believe we still have notion of base cpus and workload cpus? | 15:34 |
| kata-irc-bot | <archana.m.shinde> ok, I was playing around with that a while back, and I saw that if the default(base) cpus is set to 4 and pod requests 4 cpus, the pod gets a total of 8 cpus | 18:53 |
| kata-irc-bot | <eric.ernst> Default should be one? | 18:54 |
| kata-irc-bot | <archana.m.shinde> I wanted to look into it further, but had to look at other things. It also ties into how queues can be made available only for cold-plugged cpus | 18:55 |
| kata-irc-bot | <eric.ernst> Yeah. That's why I wanted static initially - so we can right size | 18:55 |
| kata-irc-bot | <archana.m.shinde> I want to check a few cases around this usage, i'll raise an issue/discussion once I have my thoughts on this formulated a bit more | 18:57 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!