Aye
-----Original Message----- From: Andy Allan [mailto:[EMAIL PROTECTED] Sent: 15 April 2005 10:49 To: CF-Talk Subject: Re: sorry, back to JVM/GC for a moment I've have varying success with the ConcMarkSweep collector, so yup JVM tuning is most definately a black art. On 4/15/05, Robertson-Ravo, Neil (RX) <[EMAIL PROTECTED]> wrote: > We ran an 8 month test on one of our applications (heavy use - high mem > load) and found this type of GC to be bad, very bad - there were on twin > 3ghz Xeon machines. Basically the app would run great for a while but when > it did its GC - which essentially a global GC it would pause the app > stopping connections to JRun - obviously tweaking the JVM is Application > specific *there is no optimum setup or everyone*. > > > -----Original Message----- > From: Andy Allan [mailto:[EMAIL PROTECTED] > Sent: 15 April 2005 09:38 > To: CF-Talk > Subject: Re: sorry, back to JVM/GC for a moment > > Mmmm, I need to read to emails more closely... :) > > This collector CAN be used on multiple CPU machines and just to quote > the Sun JVM docs (as I can't be botherd rehashing it myself) > > "Use the concurrent low pause collector if your application would > benefit from shorter garbage collector pauses and can afford to share > processor resources with the garbage collector when the application is > running. Typically applications which have a relatively large set of > long-lived data (a large tenured generation), and run on machines with > two or more processors tend to benefit from the use of this collector. > However, this collector should be considered for any application with > a low pause time requirement. Optimal results have been observed for > interactive applications with tenured generations of a modest size on > a single processor." > > Andy > > On 4/15/05, Andy Allan <[EMAIL PROTECTED]> wrote: > > -XX:+UseConcMarkSweep can be used successfully on single CPU machines. > > > > This collector cleans up the tenured generation in parallel with the > > execution of the application but it doesn't actually require multiple > > CPUs (although it will of course benefit from 2+ CPUs) > > > > Andy > > > > On 4/15/05, Robertson-Ravo, Neil (RX) > > <[EMAIL PROTECTED]> wrote: > > > This -XX:+UseConcMarkSweepGC should not be used on a dual processor box. > You > > > can remove it altogether to get better 0performance. > > > > > > > > > -----Original Message----- > > > From: Dov Katz [mailto:[EMAIL PROTECTED] > > > Sent: 15 April 2005 02:53 > > > To: CF-Talk > > > Subject: sorry, back to JVM/GC for a moment > > > > > > Final sanity check. Please advise if I should change anything. (w2k > server > > > 2 1ghz proc with 1gb ram - sharing sql on same box) > > > > > > I have the following JVM setting, after I modified it last week. Things > seem > > > good, but what else should I be looking to optimize things? What might > be > > > wrong/inefficient here? Thanks > > > -dov > > > > > > -server -DJINTEGRA_NATIVE_MODE -DJINTEGRA_PREFETCH_ENUMS -Xms256m > -Xmx512m > > > -Dsun.io.useCanonCaches=false > > > -Xbootclasspath/a:"{application.home}/../lib/webchartsJava2D.jar" > > > -XX:MaxPermSize=128m -XX:PermSize=64m -XX:NewSize=64m > > > -XX:+UseConcMarkSweepGC > > > > -Djavax.xml.parsers.SAXParserFactory=com.macromedia.crimson.jaxp.SAXParserFa > > > ctoryImpl > > > > -Djavax.xml.parsers.DocumentBuilderFactory=com.macromedia.crimson.jaxp.Docum > > > entBuilderFactoryImpl > > > > > > > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Discover CFTicket - The leading ColdFusion Help Desk and Trouble Ticket application http://www.houseoffusion.com/banners/view.cfm?bannerid=48 Message: http://www.houseoffusion.com/lists.cfm/link=i:4:203010 Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4 Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4 Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4 Donations & Support: http://www.houseoffusion.com/tiny.cfm/54
