CoeGSS Repository allows users with a valid portal login to access and manage the published datasets for which the user is authorized to see. The purpose of CoeGSS repository is to establish a single point of access for all the data relevant to CoeGSS whether in the form of input or output for HPC applications. Apart from making the user data accessible for HPC applications, the repository allows the CoeGSS users create and manage big or small communities, share their data publicly or for private groups enforcing the CoeGSS community.
CoeGSS repository is based on CKAN, which is basically a tool for creating open data websites. In order to provide a compact guide that is relevant to CoeGSS, only some of the functionality of CKAN is going to be described here. For standard functionality that is offered by CKAN please refer to the official documentation.
In order to access this service, the user has to be registered and logged in to the CoeGSS portal. Please see the portal guide for details. Once authenticated, the user can access this service from the Services menu on the portal navigation bar as shown on Figure 1.
Figure 1: Accessing the repository
Normally, CoeGSS Repository is aware if there is an active CoeGSS session for the user. In case Repository fails to pick up the active session, the user can authenticate themselves by clicking the login button on the Repository navigation bar as shown in Figure 2.
Figure 2: Repository authentication
In order to be able to use the CKAN REST API, an API key is required that is unique to each user. The API key can be found in the user profile page which can be accessed by clicking the user profile icon on the CKAN navigation bar as shown in Figure 3. The CKAN API key is a crucial part of the HPC Job Submission.
Figure 3: CKAN user profile access
The API key is located in the lower left corner of the user profile page as shown in Figure 4.
Figure 4: CKAN API Key
Dataset relations is an extension for CKAN that is implemented by CoeGSS. This extension allows the user to set the following relationships between the existing datasets:
The functionality of the Dataset Relations extension can be accessed by:
Figure 5: Accessing Dataset Relations Extension
In order to create a relationship between two datasets, once on the Dataset Relations tab:
Figure 6: Add/remove relation
Figure 8: Selecting the relationship type
Figure 9: Adding the relation
Existing relationships can be seen by clicking the same Add/Remove Relation button as seen in Figure 6
Existing relationships can be removed by selecting them and then clicking on the Remove relationship(s) as seen in in Figure 10
Figure 10: Removing a relationship