AMD EPYC Performance Testing with systemd

Re:AMD EPYC Performance Testing with systemd by newnewpdro on Thursday July 12, 2018 @07:31PM (#17519465)

Implying that systemd is somehow the root-cause for this performance disparity strikes me as ridiculous.

I've noticed a pattern over the years with anyone spelling systemd as SystemD: They tend to not really know what the hell they're talking about with regards to systemd, while possessing significant bias against the project, actively searching for reasons to disparage it.

[ Parent | on HN ]

Re:AMD EPYC Performance Testing with systemd by snorrah on Thursday July 12, 2018 @08:16PM (#17519655)

Feel free to illuminate us with what the possible performance discrepancy is instead caused by ?

[ Parent | on HN ]

Re:AMD EPYC Performance Testing with systemd by viraptor on Thursday July 12, 2018 @09:04PM (#17519824)

Could be phrased better, but they're not wrong: https://news.ycombinator.com/item?id=17518518

[ Parent | on HN ]

Re:AMD EPYC Performance Testing with systemd by newnewpdro on Thursday July 12, 2018 @09:50PM (#17519973)

A kernel bug, that systemd exposes by attempting to utilize more of the kernel's newer features.

[ Parent | on HN ]

Re:AMD EPYC Performance Testing with systemd by rleigh on Saturday July 14, 2018 @03:01PM (#17531791)

Possibly, but it might simply be that systemd using the shiniest new features just for the sake of it is not the most performant or sensible approach.

[ Parent | on HN ]

Re:AMD EPYC Performance Testing with systemd by jonathonf on Saturday July 14, 2018 @03:58PM (#17532055)

When running Percona under systemd the EPYC system was getting 24% _more_ throughput.

[ Parent | on HN ]