100 New Java Performance Tuning Tips

Discussions

News: 100 New Java Performance Tuning Tips

  1. 100 New Java Performance Tuning Tips (13 messages)

    Java Performance Tuning author Jack Shirazi attended 8 J1 performance sessions and published a big bulleted list with 100 new insights gained from these talks, from trivial stuff like 'use lazy loading' to cool stuff like 'Chose optimistic concurrency for CMPs with read-mostly access' and 'in servlets, set content length; for JSP pages, set UseOutputStreamSize'.

    Links to the talk PDF's are also provided.

    Check out Shirazi's big bulleted list.

    Threaded Messages (13)

  2. Huh[ Go to top ]

    This hyperlink brings us to a page full of relevant hyperlinks. Could somebody please provide a more specific link to the document in question?
  3. Huh[ Go to top ]

    The page displays 8 links that each end with ".pdf". This is not completely by accident: The author attended 8 J1 performance sessions. Thus, those links might be the "more specific links" you are looking for. Might be worth trying them out ...
  4. Huh[ Go to top ]

    Colon, under each link is a bunch of bullet points. This is the content I'm talking about.
  5. Thanks[ Go to top ]

    OK, thanks Floyd, just wanted to be sure...
  6. fails to mention...[ Go to top ]

    jvmstat: non-intrusive vm inernals monitoring tool:
    http://developers.sun.com/dev/coolstuff/jvmstat/

    how to tune gc in 1.4.2: http://java.sun.com/docs/hotspot/gc1.4.2
  7. forbidden[ Go to top ]

    All i get is

    Forbidden
    You don't have permission to access /news/newtips031.shtml on this server.

    --------------------------------------------------------------------------------

    Apache/1.3.27 Server at www.javaperformancetuning.com Port 80
  8. forbidden[ Go to top ]

    Hi Adam,

    Not sure what the issue is but we are looking into it on this end. FYI, I am remote from the site and have been able to bring it up. FYI, the relevent link is http://www.javaperformancetuning.com

    Kirk
  9. Forbidden[ Go to top ]

    I can access the site now thanks Kirk

    Adam
  10. Strange results[ Go to top ]

    I'm not quite sure about the performance results on slide "CMP vs. JDBC' within the presentation "Optimizing EJB Performance in High-Volume Data-Warehousing Applications".

    Does the vertical scale show the transactions per second, the response time, or what? In either case the results are _very_ strange. We used to do a project where we started implementing the data access with entity beans. We switched to JDBC calls after enormous performance problems. The result: data access with JDBC was _50_ (yes your read it correctly!) times faster than entity beans.

    Regards,
        Dirk
  11. Meaningless charts[ Go to top ]

    Yes, there are a number of their charts that are made totally meaningless through a lack of scale, origin, axis labelling, metric description, etc. They are not going to win any awards for presentation of data. I particularly like the ones where you cannot even tell what point they are trying to make - like that CMP/JDBC example.
  12. Strange results[ Go to top ]

    Were you using CMP v2 or v1? Everybody says CMP 2.0 is the best thing since sliced bread...
  13. RE: Strange results[ Go to top ]

    Were you using CMP v2 or v1? Everybody says CMP 2.0 is the best thing since sliced bread...


    We were using CMP 2.0 on BEA WebLogic 6.1 SP3. I'm currently evaluating a JDO implementation from Lido, since everybody says JDO is better than sliced bread :-)

    Regards,
        Dirk
  14. RE: Strange results[ Go to top ]

    When you're done with JDO check out Hibernate which is better than sliced _or_ unsliced bread. :-)