Ralph Goers

[Apache Log4j™ @ApacheLog4j]

We've just released Apache Log4j 2.19.0! This release introduces support for SLF4J 2.0 amongst other items detailed in https://t.co/RLnLBDPRhQ

Support the Log4j project via sponsoring developers via @github sponsors! https://t.co/U1LMMvLwz8

#log4j #log4j2

[Ralph Goers @rgoers]

@yazicivo @TheASF @jvz @RemkoPopma The only thing you should be surprised at is how long we took. You earned it!

[Ralph Goers @rgoers]

Log4j 1 Compatibility in Log4j 2 - https://t.co/cwej1LO6y3

[Ralph Goers @rgoers]

Why Log4j 2 was created - https://t.co/NSt71Y5jAq

[Ralph Goers @rgoers]

@RemkoPopma @qos_ch @ceki See the last 2 comments in https://t.co/fo1YPmnI4p. Converting the API to use Messages was suggested almost 10 years ago.

[Ralph Goers @rgoers]

@ceki I'm very confused. Here are my numbers for Logback 1.2. They show Logback 1.2 is about 3 times slower than Logback 1.1.10.

[Ralph Goers @rgoers]

@ceki @RemkoPopma Your numbers still show logback as slightly worse then Log4j 1. Why is that? Interesting also that RAF is so low for you.

[Ralph Goers @rgoers]

@ceki @RemkoPopma I ran again on a Linux VM I have in our prod environment. It only has 2 cores available. No idea what other VMs are doing.

[Ralph Goers @rgoers]

@ceki @RemkoPopma logback - 1.6%, log4j2 - 3.4%, log4j2 RAF - 3.7% variances. Seems an ok trade off for the performance gain.

[Ralph Goers @rgoers]

@ceki @RemkoPopma I ran it again with the same parameters as yours. I am using a Macbook Pro with 4 hyperthreads. I see you are on Windows.

[Ralph Goers @rgoers]

@RemkoPopma @ceki I ran the benchmark for Log4j 2.7 and got

[Ralph Goers @rgoers]

@connolly_s @GaryGregory @struberg @nipafx @bmathus It is difficult to make out what you are doing on twitter. Please send to the list.

[Ralph Goers @rgoers]

@kentpawz I don't know who you are and have never called you.