Re: Cek's new log4j vs logback benchmark

2021-08-27 Thread Ron Grabowski
Follow-up to "Formatting the date is expensive. The process of actually formatting a value is reasonable". Is this still an issue from LOG4J2-930: %m %ex%n: 1,755,573 msg/sec%d %m %ex%n: 1,194,184 msg/sec If so, isn't date rendering essentially a sequence we can generate ahead of time similar to

Re: Cek's new log4j vs logback benchmark

2021-08-27 Thread Carter Kozak
Thanks, Remko. The default '%d' uses FixedDateFormat with FixedFormat.DEFAULT. The FastDateFormat alternative does not support microseconds, so it doesn't suffer from the same problem. I think I can substantially reduce the frequency we re-format dates by checking FixedFormat.secondFractionDigit

Re: Cek's new log4j vs logback benchmark

2021-08-27 Thread Remko Popma
I remember looking at PatternLayout performance, I reported my findings here, hopefully they’re still useful: https://issues.apache.org/jira/browse/LOG4J2-930 If %d is used in the pattern, does the FixedDateFormat get used? > On Aug 28, 2021, at 4:33, Ralph Goers wrote: > > All of that agr

Re: Cek's new log4j vs logback benchmark

2021-08-27 Thread Ralph Goers
All of that agrees with my observations as well. Ralph > On Aug 27, 2021, at 12:23 PM, Carter Kozak wrote: > > I've identified a few things that seem impactful, but bear in mind that I > haven't begun to drill down into them yet. I plan to file individual tickets > and investigate in greater

Re: Cek's new log4j vs logback benchmark

2021-08-27 Thread Carter Kozak
I've identified a few things that seem impactful, but bear in mind that I haven't begun to drill down into them yet. I plan to file individual tickets and investigate in greater depth later on: 1. Formatting the date is expensive. The process of actually formatting a value is reasonable, howeve

Re: Cek's new log4j vs logback benchmark

2021-08-27 Thread Carter Kozak
My pleasure, I enjoy digging into this sort of performance comparison, the toughest piece right now is balancing investigation with active projects at work. I completely agree about the importance of getting this resolved quickly and I intend to continue investigating. Re loom: Currently we have a

Re: Cek's new log4j vs logback benchmark

2021-08-27 Thread Volkan Yazıcı
Carter, thanks so much for your great effort and attention on this issue. *I personally find this the uttermost important thing the project needs to address with urgency right now.* Ceki is warming up for a new Logback (and SLF4J?) release. I am pretty sure he will do his part in PR, particularly i