Description
A tool to extract captured provenance information from the Taverna provenance database. Although Taverna offers a prototype support to export captured provenance information into Janus format, we have developed an extended Provenance Extractor. The default Janus exorter plugin is incomplete and does not fully comply with the Janus ontology. As such, many object properties such as the relation between port and port value are defined as annotation, which have however no semantics and are therefore ignored by reasoning - they can not be used to infer new information. This tool maps additionally maps the gathered information to the Context Model.
Example
The left figure presents a simple Taverna 2.4 workflow, which uses a global weather web service to request the current weather information for a specific city. The right figure visualise the ontology instance of the extracted provenance workflow run.
Download & Licensing
This tool is licensend under the Apache License Version 2.0. Show License
The Performance Extractor v1.2 command line tool is available for download.
Requirements & Execution
Runtime Dependencies
- JRE 1.7
- Apache Maven 3
In the README you can find further informations about runtime dependencies.
-c,--create <path\to\create\ontology.owl> extracts performance information and creates a new ontology.
override already existing file.
-d,--database <path\to\database> change database to look up.
-f,--format <arg> ontology format. {'OWLXMLOntologyFormat'
(default),'RDFOntologyFormat','ManchesterOWLSyntax'}
-id,--runId <workflow run id> skip listing of workflows.
enter workflow id directly. (e.g.:aef30598-455a-4889-9c34-99fa520af729)
-r,--ranking <workflow run ranking> skip listing of workflows. fetch database entries ranked by execution
timestamp descending. use positiv integer (starting by 0!).
-u,--update <\path\to\existing\ontology.owl> extracts performance information and updates an existing ontology.
-v,--verbose turn on verbose mode.
In the following we present two sample executions:
java -jar target\provenanceextractor-1.2-final.jar -c \path\to\myPerformanceOntology.owl -v -f RDFOntologyFormat ===== [ID] | [Workflow Name (Timestamp)] ===== | 3 entries found. [0] Uses_the_ImageMagick @ 2014-03-19 13:11:11.398 [1] MusicClassificationExperiment @ 2014-03-13 15:11:10.287 [2] Uses_the_ImageMagick @ 2014-03-13 11:49:07.706 Your choice: //Enter an ID to process
java -jar target\provenanceextractor-1.2-final.jar -u \path\to\updatePerformanceOntology.owl -r 2 INFO - Choose worklow run (id='c0a7b302-82d5-43c7-bcd3-d315d8cd576a') by ranking number 2. // program execution proceeds