Update Reference Architecture Puma worker guidance in Docs
Further details
Proposal
Omnibus now calculates Puma worker count based on available CPU and Memory. As such for most users they no longer need to adjust these settings in installs, including with the Reference Architectures.
Currently with the Reference Architectures we advise the following:
In our architecture, we run each GitLab Rails node using the Puma webserver, and have its number of workers set to 90% of available CPUs, with four threads. For nodes running Rails with other components, the worker value should be reduced accordingly. We’ve determined that a worker value of 50% achieves a good balance, but this is dependent on workload.
Based on latest functionality this guidance can now be removed.
Who can address the issue
Other links/references
Relates !45116 (comment 430129409)
Edited by Grant Young