Tomcat 4.1.29v5.0.18 performance
I've just got my app load balanced between Apache 1.3 2xTomcat 4.1.29, JK, SunJDK1.4.2, all seperate boxes. I have set heap site (JVM settings) same as my Tomcat 5.0.18, and guessed minProcessors=100 and maxProcessors=300 for server.xml's - config different to 5.0.18. It's appears about 3x as slow as a single 5.0.18. I have three questions re this: 1) Has any1 experienced this and found a solution? 2) If I can't get T4.1.29 up2speed, and want to loadbalance with T5.0.18 / Apache 2, do people actually use this in production without any major issues? 3) If I go point 2, then the only way I've found to provide loadbalancer redundancy for those versions is LinuxVirtualServer. Is that correct? Any comments would be appreciated. Pete. *** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This footnote also confirms that this email message has been swept by MIMEsweeper for the presence of computer viruses. *** For any information on the Quinn Group of Companies please visit :- http://www.quinn-group.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Tomcat 4.1.29v5.0.18 performance
Pete Stokes wrote: I've just got my app load balanced between Apache 1.3 2xTomcat 4.1.29, JK, SunJDK1.4.2, all seperate boxes. I have set heap site (JVM settings) same as my Tomcat 5.0.18, and guessed minProcessors=100 and maxProcessors=300 for server.xml's - config different to 5.0.18. It's appears about 3x as slow as a single 5.0.18. I have three questions re this: 1) Has any1 experienced this and found a solution? 2) If I can't get T4.1.29 up2speed, and want to loadbalance with T5.0.18 / Apache 2, do people actually use this in production without any major issues? 3) If I go point 2, then the only way I've found to provide loadbalancer redundancy for those versions is LinuxVirtualServer. Is that correct? Any comments would be appreciated. If you're comparing with 5.0.18 with the HTTP connector, I'd like to point out that the HTTP connector has a higher max throughtput than using AJP with a native webserver (esp if TC and the native webserver are running on the same server). If you want to compare the two releases, you should probably be using the same configurations ;) -- x Rémy Maucherat Developer Consultant JBoss Group (Europe) SàRL x - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Tomcat 4.1.29v5.0.18 performance
Remy, With just 1 user, it appears 3x as slow. I understand the max throughput will be better, but did not expect this performance degredation with just 1 user, I'd assumed slight degredation overall and a lower max users. Don't suppose you could answer my questions 2 3 ? Thanks, Pete. Remy Maucherat wrote: Pete Stokes wrote: I've just got my app load balanced between Apache 1.3 2xTomcat 4.1.29, JK, SunJDK1.4.2, all seperate boxes. I have set heap site (JVM settings) same as my Tomcat 5.0.18, and guessed minProcessors=100 and maxProcessors=300 for server.xml's - config different to 5.0.18. It's appears about 3x as slow as a single 5.0.18. I have three questions re this: 1) Has any1 experienced this and found a solution? 2) If I can't get T4.1.29 up2speed, and want to loadbalance with T5.0.18 / Apache 2, do people actually use this in production without any major issues? 3) If I go point 2, then the only way I've found to provide apache loadbalancer redundancy for those versions is LinuxVirtualServer. Is that correct? Any comments would be appreciated. If you're comparing with 5.0.18 with the HTTP connector, I'd like to point out that the HTTP connector has a higher max throughtput than using AJP with a native webserver (esp if TC and the native webserver are running on the same server). If you want to compare the two releases, you should probably be using the same configurations ;) *** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This footnote also confirms that this email message has been swept by MIMEsweeper for the presence of computer viruses. *** For any information on the Quinn Group of Companies please visit :- http://www.quinn-group.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Tomcat 4.1.29v5.0.18 performance
Pete Stokes wrote: Remy, With just 1 user, it appears 3x as slow. I understand the max throughput will be better, but did not expect this performance degredation with just 1 user, I'd assumed slight degredation overall and a lower max users. Ok, but did you test TC 5 with the same configuration using JK ? You could be running in a performance bug when using that connector, related to whatever you are doing in your application. Don't suppose you could answer my questions 2 3 ? 2: I don't know ;) the JBoss websites aren't using it yet (since they're using JBoss 3.2.3) 3: Same. -- x Rémy Maucherat Developer Consultant JBoss Group (Europe) SàRL x - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
