Is there a benefit when having VM to changing the disk scheduler to noop on Graylog?
That depends on the kind of virtual machine.
In other words, try it and share your findings with us!
Will do, we have a single Graylog instance with 14 cores that can all be maxed out, I’m hoping this will let us process a few more logs a second when a flood of messages hits (5000/s+).
The entire server is backed by 2 x 8Gbps FC with the datastore on an all flash SAN.
Hoping this will reduce CPU load a little bit based since the system will be doing less optimization with the data being written to disk.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.