> > Just in case you missed some debugging knobs: can you retest with > > RELENG_7? (not -current). I will, but I did disable INVARIANTS/WITNESS. I was afraid I had missed something else that I am not aware of; diffing the kernel conf did not indicate this was the case. > Also for Peter: can you try different schedulers? The stock RELENG_7 > scheduler is 4BSD (reliable but not that fast), while ULE provides an > *immense* improvement. I tried both 4BSD and ULE. However, this was on a single-core machine. Both ULE and 4BSD were similar on 7, but both much worse than 6.2 with 4BSD. But alright. Sounds like I should be seeing good performance. I will re-run benchmarks with RELENG_7 and report back. If I still see unexpectedly poor performance I'll make sure to report much more detail. I will also put more effort into duplicating the exact nature of the benchmarks previously reported (by Kris IIRC). In this case the MySQL setup was not exactly equivalent because there were settings that were seemingly irrelevant and also not desired in my use case. Thank you for the update. -- / Peter Schuller PGP userID: 0xE9758B7D or 'Peter Schuller <peter.schuller_at_infidyne.com>' Key retrieval: Send an E-Mail to getpgpkey_at_scode.org E-Mail: peter.schuller_at_infidyne.com Web: http://www.scode.org
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:19 UTC