• Post Reply Bookmark Topic Watch Topic
  • New Topic
programming forums Java Mobile Certification Databases Caching Books Engineering Micro Controllers OS Languages Paradigms IDEs Build Tools Frameworks Application Servers Open Source This Site Careers Other Pie Elite all forums
this forum made possible by our volunteer staff, including ...
Marshals:
  • Campbell Ritchie
  • Jeanne Boyarsky
  • Ron McLeod
  • Paul Clapham
  • Liutauras Vilda
Sheriffs:
  • paul wheaton
  • Rob Spoor
  • Devaka Cooray
Saloon Keepers:
  • Stephan van Hulst
  • Tim Holloway
  • Carey Brown
  • Frits Walraven
  • Tim Moores
Bartenders:
  • Mikalai Zaikin

Neophyte app performance question

 
Greenhorn
Posts: 2
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
<disclaimer>I am not a developer</disclaimer>

I have a situation wherein an application requires approximately a 2gig heap to function. Garbage collection at times can take upwards of 7seconds and will cause failures in this application. We are not paging, all other server performance statistics are so low that they are inconsequential. The admin is telling me that there is nothing else that he can do to make the application perform better. Is there nothing else we can do to tune/modify the way that garbage collection is happening? I have suggested splitting the app across several JVM's to reduce the total required heap and the time it takes to do full garbage collection. There may be issues with this he says (due to the way the app functions). That said - any other ideas where I should push/prod him?

This is a packaged application - so we cannot modify the source - and he is running it within a sun JVM.

Thanks for your insight and patience.

Sanjay
 
Ranch Hand
Posts: 83
Spring Tomcat Server Java
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
2Gig is a huge memory ....What is this application doing? Are you keeping lots of data in memory ?

You may need to give more details on the kind of application you are using.

Also tell following
1. What is machine configuration? RAM, Processor Disk space etc.
2. What is the java min/max heap size of this application.

If you application is crashing due to huge memory requirements then there could be two possible causes
1. Either your application has a memory leak or
2. It really requires that much of memory and it is not able to manage it by doing swapping also.
 
author & internet detective
Posts: 41860
908
Eclipse IDE VI Editor Java
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
Sanjay,
Welcome to JavaRanch!

Where did the application come from? Can you ask the vendor for help?
 
author
Posts: 14112
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
What kind of failures is this causing?

Did you take a look at http://java.sun.com/docs/hotspot/gc5.0/gc_tuning_5.html ?
 
author
Posts: 5856
7
Android Eclipse IDE Ubuntu
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
Have you gather gc data and looked at it to see what is really going on? Which collector are you using - would the concurrent collector benefit you? See my presentation at http://www.cecmg.de/doc/tagung_2007/agenda07/24-mai/2b3-peter-johnson/index.html
 
sanjay ramaswamintahan
Greenhorn
Posts: 2
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
I am going to work with the admin to collect some gc perf stats. The vendor is throwing up their hands stating that we are running out of spec. We will try the concurrent GC model as well and see if that has any measurable benefit. Many thanks to all of you for your responses!

-sanjay
reply
    Bookmark Topic Watch Topic
  • New Topic