http://www.basilv.com/psd/blog/2010/avoiding-caching-to-improve-hibernate-performance Wed, 16 May 2012 04:03:57 +0000 hourly 1 http://wordpress.org/?v=3.3 http://www.basilv.com/psd/blog/2010/avoiding-caching-to-improve-hibernate-performance/comment-page-1#comment-108907 James Cronen Mon, 14 Jun 2010 19:14:19 +0000 http://www.basilv.com/psd/?p=486#comment-108907 Great post! Thank you for this; you've gotten me out of a serious performance jam. Great post! Thank you for this; you’ve gotten me out of a serious performance jam.

]]> http://www.basilv.com/psd/blog/2010/avoiding-caching-to-improve-hibernate-performance/comment-page-1#comment-105113 Angel Sun, 07 Mar 2010 09:18:31 +0000 http://www.basilv.com/psd/?p=486#comment-105113 Hi. I still think that in a batch-like scenario the first level cache is for no use. Anyway you always have the choice of using whatever Hibernate session that is more suitable for your needs, and you can use the normal Hibernate Session for the rest of the application. But I would never use the normal Hibernate session for a batch process like the one you describe, or you will end up having severe performance issues like those you have mentioned in your post. I still think the best solution to that problem is as easy as using the stateless session, with no drawbacks at all. Cheers. Hi.
I still think that in a batch-like scenario the first level cache is for no use. Anyway you always have the choice of using whatever Hibernate session that is more suitable for your needs, and you can use the normal Hibernate Session for the rest of the application. But I would never use the normal Hibernate session for a batch process like the one you describe, or you will end up having severe performance issues like those you have mentioned in your post. I still think the best solution to that problem is as easy as using the stateless session, with no drawbacks at all.
Cheers.

]]> http://www.basilv.com/psd/blog/2010/avoiding-caching-to-improve-hibernate-performance/comment-page-1#comment-104992 Basil Vandegriend Thu, 04 Mar 2010 05:04:04 +0000 http://www.basilv.com/psd/?p=486#comment-104992 Using stateless Hibernate sessions is an option, but since such sessions have no level one cache at all, you lose out on a lot of Hibernate functionality that I consider essential (see https://www.hibernate.org/hib_docs/v3/api/org/hibernate/StatelessSession.html for details). So I consider stateless sessions to be an option of last resort. You are correct that filling the level one cache with thousands of objects hurting performance is poor usage of Hibernate. Using stateless Hibernate sessions is an option, but since such sessions have no level one cache at all, you lose out on a lot of Hibernate functionality that I consider essential (see https://www.hibernate.org/hib_docs/v3/api/org/hibernate/StatelessSession.html for details). So I consider stateless sessions to be an option of last resort. You are correct that filling the level one cache with thousands of objects hurting performance is poor usage of Hibernate.

]]> http://www.basilv.com/psd/blog/2010/avoiding-caching-to-improve-hibernate-performance/comment-page-1#comment-104946 Angel Wed, 03 Mar 2010 06:06:55 +0000 http://www.basilv.com/psd/?p=486#comment-104946 Correct me if I am wrong, but in the scenario you are mentioning (batch process) you should be using Stateless Hibernate sessions. If not is normal to get those performance figures. So I think in this case is not a problem with the cache but a misuse of Hibernate. Correct me if I am wrong, but in the scenario you are mentioning (batch process) you should be using Stateless Hibernate sessions. If not is normal to get those performance figures. So I think in this case is not a problem with the cache but a misuse of Hibernate.

]]>