Group items tagged

Advanced usage Input arrays An application might require arrays of inputs, whose exact sizes are specified by the user when creating the job. This is supported by GRIA using the minOccurs, maxOccurs and defaultSize attributes on DataStager elements. For example, if your application took between 2 and 8 images as input, you might use the following XML: <DataStager type="input" name="inputImage" minOccurs="2" maxOccurs="8" defaultSize="2"> <Description>Input image</Description> <MimeType>image</MimeType></DataStager> You can use the defaultSize attribute to support older clients that do not know how to specify the desired size of arrays.

Advanced usage Input arrays An application might require arrays of inputs, whose exact sizes are specified by the user when creating the job. This is supported by GRIA using the minOccurs, maxOccurs and defaultSize attributes on DataStager elements. For example, if your application took between 2 and 8 images as input, you might use the following XML: <DataStager type="input" name="inputImage" minOccurs="2" maxOccurs="8" defaultSize="2"> <Description>Input image</Description> <MimeType>image</MimeType></DataStager> You can use the defaultSize attribute to support older clients that do not know how to specify the desired size of arrays.

Advanced usage Input arrays An application might require arrays of inputs, whose exact sizes are specified by the user when creating the job. This is supported by GRIA using the minOccurs, maxOccurs and defaultSize attributes on DataStager elements. For example, if your application took between 2 and 8 images as input, you might use the following XML: <DataStager type="input" name="inputImage" minOccurs="2" maxOccurs="8" defaultSize="2"> <Description>Input image</Description> <MimeType>image</MimeType></DataStager> You can use the defaultSize attribute to support older clients that do not know how to specify the desired size of arrays.

Running a Groovy script   This is the easiest way to access the Java API. To try it out, create a script named test.groovy containing: println "Starting Groovy script!" def jobService = serviceFactory.createServiceProxy("https://host:8443/gria-basic-app-services/services/JobService?wsdl") println "Getting ID..." def id = jobService.getServiceProviderID() println "Job service identity: " + id Replace host with the address of your GRIA job service. The script is run by passing it as an argument to the normal gridcli command. For example: $ gridcli test.groovy Starting Groovy script! Getting ID... Job service identity: <subject> <subject> will appear at the identity of your job service. The configured keystore and active plugins are the same as when running the GRIA client normally.  

The GRIA Workflow Application software includes tools for deploying and running Taverna workflows as GRIA applications. A command line deployment tool generates GRIA application wrapper scripts and other requisite files from a Taverna workflow. Running the deployment tool generates a new application that can be deployed to the GRIA Job service in the usual way using the Job Service Administration web pages.

The GRIA Workflow Application v0.2 software includes tools for deploying and running XScufl workflows as GRIA applications. XScufl workflows may be created using Taverna or g-Eclipse. A command line deployment tool generates GRIA application wrapper scripts and other requisite files from an XScufl workflow. Running the deployment tool generates a new application that can be deployed to the GRIA Job service in the usual way using the Job Service Administration web pages.

For Linux, the first line of the script (eg. #!/usr/bin/python) is used to determine which interpreter to use. The filename extension can be anything (eg. startJob.py, startJob.sh).

The application wrapper scripts are deployed along with the applications - either on a shared file system or individually on each compute node. Their function is to provide a uniform interface to the Job Service for starting, monitoring and stopping applications.

This guide describes the GRIA Workflow Deployer Application software which works in conjunction with the GRIA Workflow Application package, and enables the automatic conversion of a user-supplied XScufl workflow into a GRIA application, which is then automatically deployed onto a GRIA Job Service. XScufl workflows may be created using Taverna or g-Eclipse.

The bulk of the changes in 5.3 are to the client API. One exciting new > feature is built-in support for Groovy, a scripting language with a > Java-like syntax. For example, the following script will upload the > "source.jpg" file, process it with the "paint" application on > griademo1 > and then download the result:

> def GRIADEMO1_JOB_SERVICE_WSDL = > "https://griademo1.it-innovation.soton.ac.uk/gria-basic-app-services/services/JobService?wsdl > > " > def JOB_TYPE = "http://it-innovation.soton.ac.uk/grid/imagemagick/paint > " > def jobService = > serviceFactory.createServiceProxy(GRIADEMO1_JOB_SERVICE_WSDL) > def swirlJob = jobService.createJob(JOB_TYPE, "My swirl job") > swirlJob.input("inputImage").saveFromFile(new File("source.jpg")) > swirlJob.startJob(null) > while (swirlJob.stillActive()) { > println "Waiting..." > Thread.sleep(1000) > } > swirlJob.output("outputImage").read(new File("result.jpg")) > swirlJob.destroy() > > To try it, save the script as "paint.groovy" and then run the client > like this (you'll need a "source.jpg" file too, of course): > > $ gridcli ./paint.groovy > > The above script also demonstrates the new API. "serviceFactory" takes > the URL of a service's WSDL and creates a proxy for it. Notice that, > unlike the 5.2 API, 5.3 does not require you to store the proxy in a > repository.

This allows you to access managed Basic Application Services on griademo2. You can now use the client to store data and run jobs, much as you did before, but managed by the SLA that you proposed and billing to the trade account that you have just opened:

Use nested workflows (sub workflows)

Monitor execution

Manage jobs and results