Skip to end of metadata
Go to start of metadata

It is possible for Taverna Workbench to run out of memory, for example when you are running a workflow that is handling large amounts of data or a large number of calls to services.  This problem is normally indicated by a:

in either Taverna's log file, in the command window that you are running the workbench from, or in a popped-up window.

The first thing to check is your data and provenance preferences. If you are using In-memory storage then you should untick this option so that Taverna Workbench stores data in a database.  This will enable Taverna Workbench to handle more data.

If you still have problems, then you can change the amount of memory allocated to running the Taverna Workbench.

Before you change the Taverna Workbench's memory allocation you need to shut down the Taverna Workbench.

Windows

To change the memory allocation you will need to edit taverna.vmoptions file inside Taverna’s installation directory (this may require administrator privileges) and replace -Xmx1g with say -Xmx1500m to use 1.5 Gigabytes as below:

When you start Taverna next time (using taverna.exe or the start menu) it should have an increased memory allocation.

If you are using the platform-neutral Taverna on Windows or for some other reason start Taverna using taverna.bat instead of taverna.exe you would have to modify taverna.bat file in a similar fashion.

Mac OS

Go to where your Taverna application is installed (normally the Applications folder), right-click on Taverna and select Show Package Contents. You need to edit the file Info.plist in the Contents folder. Right-click on Info.plist and Open With the TextEdit program.

Scroll down to the bottom of the file and you will see a line (or similar to):

The 1g indicates that Taverna Workbench is currently allocated a maximum of 1000 Megabytes (1 Gigabytes).  Increase the value, for example to 1200m, and save Info.plist.  When you start Taverna next time it will have an increased memory allocation.

If you are using the platform-neutral Taverna on OS X – you would have to modify taverna.sh file in a similar fashion.

Linux

To change the memory allocation you will need to edit taverna.vmoptions file inside Taverna’s installation directory (this may require administrator privileges) and replace -Xmx1000m with say -Xmx1200m to use 1.2 Gigabytes on the line like below:

When you start Taverna next time using taverna (not taverna.sh) it will have an increased memory allocation.

 

If you are using the platform-neutral Taverna on Linux, or for some other reason are starting Taverna using taverna.sh instead of taverna – you would have to modify taverna.sh file in a similar fashion.
Labels
  • None