Unnecessary loading of @ContainedIn entities causes big performance hit ----------------------------------------------------------------------- Key: HSEARCH-782 URL: http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-782 Project: Hibernate Search Issue Type: Bug Components: engine Affects Versions: 3.4.0.Final Environment: Hibernate 3.6.3 MS SQL Server 2005 Reporter: Adam Harris Attachments: log4j.properties, optimizations.zip, org.hibernate.search.test.engine.optimizations.CollectionUpdateEventTest2-output.txt This bug is somewhat related to HSEARCH-679 where I have added some comments. Here is a simplified model: {code:title=A.java|borderStyle=solid} public class A { @Field private String name; @ContainedIn private Collection<B> b; } {code} {code:title=B.java|borderStyle=solid} @Indexed public class B { @Field private String name; @IndexEmbedded private A a; } {code} The basic problem is that every time a new instance of B is created and added to A's collection, all of the B instances in that collection have their index recreated. The reason this is a problem is that nothing in A has changed (e.g. the "name" field). There is no new information to add to all of the B indexes. As the number of objects in the collection grows, this becomes are real performance hit. It seemed like HSEARCH-679 was trying to address this problem, but this may be a little different scenario. I'm attaching some sample code and my log file that demonstrates the extra index events. Please see line 519 of the surefire log file. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira