DataStorage

Overview

Many communities have existing databases. They may wish to access these databases from workflows and/or store values computed within workflowas in a custom database. Although Taverna has a data store that captures data created within workflow runs, the associated mechanism does not address the situation in which the user has or would like to have a domain-specific database for use outside myGrid.

Overall Goals

The overall goal is to ease the task of accessing (reading from and writing to) domain-specific relational databases from workflows. It is assumed that the person writing the workflow has the permissions and skills required to send SQL statements to a database, or knows someone who has.

Affected Components

The expectation is that the requirement can be met by the development of a processor and associated interface. Users will be able to study the contents of an existing database, and either:

1. write queries that extract data from the database, where the query may be parameterised with values passed in from the workflow;
2. write requests that insert data from the workflow into a named table in the database.

Key Tasks

1. identify relevant use cases, for example from ISPIDER;
2. design the relevant processor and associated interface (cf BioMart);
3. implement the processor (note that in terms of database access, the same user functionality should be able to be supported both over JDBC and OGSA-DAI, with the only differences being within the code of the plugin);
4. test.

Developers

• David Withers