Research data spring

Data Vault27/2/2015

Problem

» Active data storage being used for everything

» No interface to archiving (good RDM practice)

» Archiving undertaken on unsustainable/non-compliant storage solutions (DVDs / USB drives)

» No consideration for curation and archiving, data description, retention and review

» Items of long term value not actively identified and realised

07/04/2015 Data Vault project pitch 2

Solution

07/04/2015 Data Vault project pitch 3

Use cases

» A paper has been published, and according to the research funder’s rules, the data underlying the paper must be made available upon. It is therefore important to store a date-stamped golden-copy of the data associated with the paper. Even if the author’s own copy of the data is subsequently modified, the data at the point of publication is still available.

» Data containing personal information, perhaps medical records, needs to be stored securely, however the data is ‘complete’ and unlikely to change, yet hasn’t reached the point where it should be deleted.

» Data analysis of a data set has been completed, and the research finished. The data may need to be accessed again, but is unlikely to change, so needn’t be stored in the researcher’s active data store. An example might be a set of completed crystallography analyses, which whilst still useful, will not need to be re-analysed.

» Data is subject to retention rules and must be kept securely for a given period of time, for example EPSRC funded data that needs to be stored securely for ten years.

07/04/2015 Data Vault project pitch 4

Impact and Benefits

» Impact:

› Encourage archiving and proper long term storage of data

› This leads to better data management practice

» Benefits

› Researchers

› University HE Community

› IT Services

07/04/2015 Data Vault project pitch 5

Sustainability

» Open source software developed in github

» Monthly sprints with fortnightly catchups

» Institutional commitment

» Embedding in RDM Programmes

» Solution open to all – we’ve seen a lot of interest over time, and during this sandpit event

07/04/2015 Data Vault project pitch 6

Outputs, milestones and indicators of success

» Phase 1

› Month 1– Project kick-off– Investigation Report (how to package data in containers,

common API methods from storage providers)

› Month 2 – Specification (define how the Data Vault will work)

› Month 3– Proof of concept (deliver an end to end proof of concept)

07/04/2015 Data Vault project pitch 7

Outputs, milestones and indicators of success

» Phase 2

› 4 months– Minimum viable product

»Phase 3› 6 months– Full Release

07/04/2015 Data Vault project pitch 8

Outputs, milestones and indicators of success

» Indicators of Success:

› Researchers using the tool

› Other Institutions take-up

› Freeing up of active data storage

› A move towards the culture change required

› Ability to manage and realise assets

07/04/2015 Data Vault project pitch 9

Team

» Project members:

› University of Edinburgh

› University of Manchester

› Other active contributors welcome

» Project leads on both sites (institutional contribution)

» Developers x 1 per partner, half funded by JISC, half locally funded

» Plus each sites RDM Teams07/04/2015 Data Vault project pitch 10

Funding

Jisc Local 3 months - Jisc3 months -local

RDM Developer - Edinburgh £2,500 £2,500 £7,500 £7,500

RDM Developer - Manchester £2,500 £2,500 £7,500 £7,500

3 * monthly project meetings 666 2000

Next Research Data Spring workshop 500 500

£17,500

07/04/2015 Data Vault project pitch 11

Research data spring

Data Vault – thank you!27/2/2015