You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@openwhisk.apache.org by OpenWhisk Team Slack <ra...@apache.org> on 2022/08/28 15:17:07 UTC

[slack-digest] [2022-08-27] #general

2022-08-27 07:49:45 UTC - Rajiv Vaidyanathan: Follow-up to the previous question: Even when there is an invocation to functions in warmed containers, there is a cache access at the invoker. Why is cache access in the critical path for the warm container?
https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1661586585255239?thread_ts=1661586585.255239&cid=C3TPCAQG1
----
2022-08-27 12:50:53 UTC - Kan Wang: I cannot believe that the composer parallel slower than a sequence base on my testing. Especially when the number of fan-out bigger than 5, the delay can be tens seconds to a minute. My tested environment is one node cluster locally, and the parameter size is just one or two small characters. Have any one used the composer parallel? could you share me your fan-out performance please? Thank you.
https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1661604653032339?thread_ts=1661604653.032339&cid=C3TPCAQG1
----