<HTML dir=ltr><HEAD><TITLE>Re: Tenuring distribution</TITLE>
<META http-equiv=Content-Type content="text/html; charset=unicode">
<META content="MSHTML 6.00.6000.16890" name=GENERATOR></HEAD>
<BODY>
<DIV id=idOWAReplyText66020 dir=ltr>
<DIV dir=ltr><FONT face=Arial color=#000000 size=2>The snippet below is for uniform test load in staging environment.</FONT></DIV>
<DIV dir=ltr><FONT face=Arial size=2></FONT> </DIV>
<DIV dir=ltr><FONT face=Arial size=2>Thanks,</FONT></DIV>
<DIV dir=ltr><FONT face=Arial size=2>Sujit</FONT></DIV></DIV>
<DIV dir=ltr><BR>
<HR tabIndex=-1>
<FONT face=Tahoma size=2><B>From:</B> Y.S.Ramakrishna@Sun.COM [mailto:Y.S.Ramakrishna@Sun.COM]<BR><B>Sent:</B> Mon 8/3/2009 2:27 PM<BR><B>To:</B> Das, Sujit (Cognizant)<BR><B>Cc:</B> hotspot-gc-use@openjdk.java.net<BR><B>Subject:</B> Re: Tenuring distribution<BR></FONT><BR></DIV>
<DIV>
<P><FONT size=2>On 08/ 3/09 12:01 PM, Sujit.Das@cognizant.com wrote:<BR>> Hi All,<BR>> <BR>> I am trying to look whether there is any scope in improving tenuring<BR>> distribution by changing survivor space or Max Tenuring Threshold. I<BR>> have read some Tenuring distribution examples. But following tenuring<BR>> distribution output of my application confuses me and not sure whether<BR>> the young generation is well-tuned or not. The following tenuring<BR>> distribution O/P is result of consecutive minor collections. It<BR>> increases from age 1 to 3 (sometimes upto age 4) and then drops to age<BR>> 1. It then again increases with each minor collection.<BR><BR>Yes, as you can see, the TT drops to 3 and then to 1, and this<BR>happens because there is periodically a large number of young survivors<BR>(i.e. age 1) in the survivor space, so we pro-actively promote at the<BR>next scavenge to avoid the possibility of premature overflow. You could increase your<BR>survivor space size. (There is clearly some kind of periodic generation<BR>of a large volume of data in your application. You can see whether that<BR>correlates to any kind of load-pulsing, or convoying effects. Are you<BR>using a production load (which can be bursty) or a bursty test load<BR>or a uniform test load for these figures?)<BR><BR>Once again, I think Tony & Charlie's talk slides will help in this kind of<BR>tuning.<BR><BR>-- ramki<BR><BR>> <BR>> Please do let me know your thoughts.<BR>> <BR>> Desired survivor size 14188544 bytes, new threshold 4 (max 4)<BR>> - age 1: 7040008 bytes, 7040008 total<BR>> : 305088K->13787K(305088K), 0.0395292 secs]<BR>> 906769K->626948K(3044288K)After GC:<BR>> ...<BR>> Desired survivor size 14188544 bytes, new threshold 4 (max 4)<BR>> - age 1: 6448424 bytes, 6448424 total<BR>> - age 2: 4461088 bytes, 10909512 total<BR>> : 291163K->12426K(305088K), 0.0191044 secs]<BR>> 904324K->625588K(3044288K)After GC:<BR>> ...<BR>> Desired survivor size 14188544 bytes, new threshold 3 (max 4)<BR>> - age 1: 7150744 bytes, 7150744 total<BR>> - age 2: 4394936 bytes, 11545680 total<BR>> - age 3: 3582184 bytes, 15127864 total<BR>> : 289802K->19070K(305088K), 0.0191912 secs]<BR>> 902964K->632232K(3044288K)After GC:<BR>> ...<BR>> Desired survivor size 14188544 bytes, new threshold 1 (max 4)<BR>> - age 1: 18196472 bytes, 18196472 total<BR>> - age 2: 3800224 bytes, 21996696 total<BR>> - age 3: 3340344 bytes, 25337040 total<BR>> : 296446K->27712K(305088K), 0.0282980 secs]<BR>> 909608K->648613K(3044288K)After GC:<BR>> ...<BR>> Desired survivor size 14188544 bytes, new threshold 4 (max 4)<BR>> - age 1: 5757968 bytes, 5757968 total<BR>> : 305088K->13117K(305088K), 0.0403760 secs]<BR>> 925989K->645051K(3044288K)After GC:<BR>> ...<BR>> Desired survivor size 14188544 bytes, new threshold 4 (max 4)<BR>> - age 1: 5984000 bytes, 5984000 total<BR>> - age 2: 4197480 bytes, 10181480 total<BR>> : 290493K->13133K(305088K), 0.0251218 secs]<BR>> 922427K->645067K(3044288K)After GC:<BR>> ...<BR>> Desired survivor size 14188544 bytes, new threshold 3 (max 4)<BR>> - age 1: 6765056 bytes, 6765056 total<BR>> - age 2: 4159752 bytes, 10924808 total<BR>> - age 3: 3534520 bytes, 14459328 total<BR>> : 290509K->17608K(305088K), 0.0258073 secs]<BR>> 922443K->649542K(3044288K)After GC:<BR>> ...<BR>> Desired survivor size 14188544 bytes, new threshold 1 (max 4)<BR>> - age 1: 17919056 bytes, 17919056 total<BR>> - age 2: 4852200 bytes, 22771256 total<BR>> - age 3: 3152768 bytes, 25924024 total<BR>> : 294984K->27712K(305088K), 0.0303612 secs]<BR>> 926918K->667692K(3044288K)After GC:<BR>> ...<BR>> Desired survivor size 14188544 bytes, new threshold 4 (max 4)<BR>> - age 1: 6579448 bytes, 6579448 total<BR>> : 305088K->13396K(305088K), 0.0386087 secs]<BR>> 945068K->664582K(3044288K)After GC:<BR>> <BR>> Thanks,<BR>> Sujit<BR>> This e-mail and any files transmitted with it are for the sole use of<BR>> the intended recipient(s) and may contain confidential and privileged<BR>> information.<BR>> If you are not the intended recipient, please contact the sender by<BR>> reply e-mail and destroy all copies of the original message.<BR>> Any unauthorized review, use, disclosure, dissemination, forwarding,<BR>> printing or copying of this email or any action taken in reliance on<BR>> this e-mail is strictly prohibited and may be unlawful.<BR>><BR>><BR>> ------------------------------------------------------------------------<BR>><BR>> _______________________________________________<BR>> hotspot-gc-use mailing list<BR>> hotspot-gc-use@openjdk.java.net<BR>> <A href="http://mail.openjdk.java.net/mailman/listinfo/hotspot-gc-use">http://mail.openjdk.java.net/mailman/listinfo/hotspot-gc-use</A><BR><BR></FONT></P></DIV></BODY></HTML>
<table><tr><td bgcolor=#ffffff><font color=#000000>This e-mail and any files transmitted with it are for the sole use of the intended recipient(s) and may contain confidential and privileged information.<br>
If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. <br>
Any unauthorized review, use, disclosure, dissemination, forwarding, printing or copying of this email or any action taken in reliance on this e-mail is strictly prohibited and may be unlawful.<br>
</font></td></tr></table>