IJC 2.3.1 Out of Memory Error

I was building a database of structures by importing SD files. As I neared the 2 million structure level I began to receive out of memory error messages on the final stage of import but all of the structures were loaded in the grid view. After each of these events I would close the application and then restart before loading the next set. At about 2.2 million structures, I could not get IJC to load. Is there a limit on the size of the database? Is there a way to recover back to the last successful load? IJC running on PC under VISTA, 3 GB memory. Error message follows:





java.lang.OutOfMemoryError: Java heap space


at java.util.Arrays.copyOf(Unknown Source)


at java.util.Arrays.copyOf(Unknown Source)


at java.util.ArrayList.toArray(Unknown Source)


at java.util.ArrayList.<init>(Unknown Source)


at com.im.df.impl.db.ResultSetImpl$VS.updateIdsAndSelectionInternal(ResultSetImpl.java:545)


at com.im.df.impl.db.ResultSetImpl$VS.setIds(ResultSetImpl.java:534)


at com.im.df.impl.db.ResultSetImpl$VS.access$200(ResultSetImpl.java:385)


at com.im.df.impl.db.ResultSetImpl.applyQueryInternal(ResultSetImpl.java:192)


at com.im.df.impl.db.ResultSetImpl.<init>(ResultSetImpl.java:88)


at com.im.df.impl.db.DBSchemaDataProviderImpl.getDefaultResultSet(DBSchemaDataProviderImpl.java:88)


at com.im.ijc.core.api.views.AbstractViewTopComponent$8.phase1InRequestProcessor(AbstractViewTopComponent.java:828)


at com.im.df.util.BackgroundRunner.processTheTask(BackgroundRunner.java:112)


at com.im.df.util.BackgroundRunner$1.run(BackgroundRunner.java:79)


at org.openide.util.RequestProcessor$Task.run(RequestProcessor.java:561)


at org.openide.util.RequestProcessor$Processor.run(RequestProcessor.java:986)

When handling very large datasets like this you should definitely increase the amount of memory that is available to IJC.


To see how to do this look here:


http://www.chemaxon.com/instantjchem/ijc_latest/docs/user/help/htmlfiles/changing_user_settings.html#startup


and for more info about the topic see here:


http://www.chemaxon.com/instantjchem/ijc_latest/docs/user/help/htmlfiles/tips_and_tricks/memory_usage.html





For 2 million structures I would suggest at least 400MB, or more if you have it.





You can edit this value in the configuration file if you need.


Look for the file:


<ijc_installation_dir>/etc/instantjchem.conf


and look for the line starting with


default_options


and edit the value for the -J-Xmx setting to something like:


-J-Xmx400m








If your import had completed then once you increase the memory setting you should see all the data that was imported.








Tim

Thanks. Using the inforamtion provided and reading the links, I was able to change the starting and max memory settings. This solved the problem.

OK, good.


I also mention that IJC 2.4 will contain some improvements in handling very large datasets.





Tim