3.
Tutorial
Up one level
The tutorial assumes that you are using GRIA 5 services that are free (and do not require valid SLAs to be presented by the client). However, with some additional configuration of the Upload and Job processors in the tutorial workflow, you will be able to use GRIA 5 services that require SLAs. If you wish to use such services, refer to the Reference section for details of how to configure processors to use SLAs.
Overview
We'll build a workflow in Taverna that performs image processing. The first step in using the GRIA plugins will involve using Taverna to discover the applications that are available at a GRIA service provider. We'll then build the workflow for image processing. The first step in the workflow will be a processor for uploading an image file to a GRIA server. After adding the upload processor, we'll add two job processors for performing image processing computations. The final step in the workflow will be a download processor for retrieving the output image file from the remote GRIA server.
The final workflow will look like the diagram below.
Prerequisites
You should be already familiar with the GRIA client and have completed the Client User's Tutorial.
You should be familiar with constructing and executing workflows using Taverna. If not, please refer to the Taverna documentation.
In addition, the GRIA Workflow Plugins and prerequisite software must have been installed and configured according to the instructions in the Installation section.
Start Taverna
On WinXP, start the Taverna workbench by double clicking on the file:
TAVERNA_HOME/runme.bat
On Linux systems, first change directory to TAVERNA_HOME. Next, set execute permissions on the runme.sh, before executing it.
chmod u+x runme.sh
./runme.sh
Discover applications
Discover the applications that are available at your chosen GRIA service provider.
In the Available services panel tree, right-click on the root node, named Available Processors. In the context menu that's displayed, select Add applications from a GRIA 5.* service provider, as seen below.
Discover applications
You will be asked to enter the Address (URL) of the GRIA service provider. Enter this appropriately. Usually this will involve simply replacing the HOST:PORT placeholder in the text field with the name of the server. You may also need to change the URL prefix from https to http if the services aren't secured by SSL. Alternatively you can select URLs you have previously used.
After entering the address, the system will query the remote server to discover details of the applications that are available. At this point you may be warned that the certificate presented by the remote service is issued by an unknown certificate authority. This means that there is no way to check that it is the server it claims to be. In this case, and when you are using the software for evaluation and testing purposes only, you can select that you wish to trust the certificate temporarily for this session.
After discovering applications, the service providers' applications will be represented as nodes (Processors) in the tree in the Available services panel. These can be added as Processors in workflows.
Compose workflow
Add an upload processor
Add an upload processor to transfer the input image file from your machine to the remote server. Expand the node in the Available services panel tree that represents the GRIA server provider. Under the Data node right click on the Upload node and in the context menu select Add to model.
Add a paint job
Under the
Applications node right click on paint application node,
named:
http://it-innovation.soton.ac.uk/grid/imagemagick/paint
and in the context menu select Add to model with name.... Call the new processor paint.
Add a paint job processor
Add a swirl job
As done above for paint, add a swirl processor to the workflow. The swirl application node is called:
http://it-innovation.soton.ac.uk/grid/imagemagick/swirl
but choose a more readable name such as swirl.
Add a download processor
Add a download processor to retrieve the results file from the remote server. The download
processor is available under the
Data node for the GRIA service provider.
Add a download processor
Create data links
Place data links between processors in the workflow until the workflow looks like the workflow diagram at the start of this section. Recall from the Taverna documentation that data links are added using context menus available from the Advanced model explorer panel. As an example, consider making the data link from paint to swirl. Select the painted-image port of the paint processor in the Advanced model explorer panel. Right-click and use the context menu to make a data link to the source-image input port of the swirl processor. Add the other data links in a similar way.
Add data links
Bind inputs
Specify file names for the input and output image files. The workflow is complete except for specifying the file to upload and what to call the results file. Provide these values by editing the localFile input ports. For each String, right-click on the node in the Advanced model explorer panel. Select Set default value from the context menu and provide a name for the file in the dialog that's displayed. Note that an absolute file name can be used. Alternatively, a relative file name may be used and in this case the path is relative to TAVERNA_HOME.
Specify filenames
Your workflow should now look similar to the workflow at the start of this section. It will have input data bound and therefore will be ready to execute.
Execute workflow
Run the workflow by selecting Run workflow from the Tools and workflow invocation menu.
Run the workflow
View results
If the workflow completes successfully, the workflow status panel should look similar to the figure below and the resulting image file should be available on your file system at the location you specified.
Successful workflow completion
Next steps
In this tutorial, the default GRIA Upload and Job processor settings have been used. To understand how the processors can be configured, please read the Reference section.
