Agenda:
- Introductions (5 minutes)
- Brief overview of MDF Project (10 minutes)
- Brief Background on 4CeeD Project (5 minutes)
- Discuss potential benefits of 4CeeD/MDF Integration (10 minutes)
- Candidate datasets and metadata (5 minutes)
- Current deployment architecture (10 minutes)
- Discuss combined team work practices (10 minutes)
- Agree next steps (5 minutes)
Deployment architecture
Runs on a single bare metal server as a docker image in docker-compose stack.
Server has a gluster filesystem attached. This is the file store for the mongo db.
Potential Benefits of 4CeeD/MDF Integration
- Enable increased scale for 4CeeD by offloading archival data storage and retrieval to MDF infrastructure
- Bring durability to the data curated by 4CeeD by allowing investigators to publish datasets to a long-lived repo and receiving a DOI or Handle to the data.
- MDF outreach is currently focused on "Live Data" - are there opportunities to extract more value from student datasets by publishing them as it is collected. Could students in 500 level classes create images on materials that are of interest to a wider community and make them available via MDF?
- Ability to perform Forge queries that are informed by data held in 4CeeD. Have the resulting datasets downloaded from MDF for integration.
Next Steps
Reconcile metadata between 4CeeD and MDF. We expect that the 4CeeD metadata is a superset of MDF. If that is the case, is there a way to store the full metadata in a user-defined metadata field?
Investigate integration with IDEAL, campus archive.
Review License in the extractor repo
Working Together
Edmond Chow (Sr. Research Engineer at MNTL has been the prime customer of 4CeeD. Brian Cunningham, Director of MNTL says that every grad student who uses the lab goes through Edmond.
Patrick Su is the graduate student working with 4CeeD. His data will eventually be a good candidate for publication.
Steve Konstanty is the project leader on the project and decides on priorities.