Jisc Shared Services Workshop 9 July 2015
At the recent Research Data Shared Services Workshops a draft top level RDM architecture was presented for feedback from participants. In this post John Kaye explains this architecture and its purpose and welcomes feedback from the wider community
In planning RDM shared services for Jisc’s Research at Risk co-design challenge there has been a need to draft a top level ‘ideal’ conceptual RDM architecture to help Jisc and Higher Education Institutions evaluate existing infrastructure, data and metadata flows and find out where institutions may have gaps or requirements that Jisc could assist with through RDM shared services.
The workshop was presented with a draft Jisc top level RDM Infrastructure diagram, that has been based on the concepts developed by John Lewis from University of Sheffield and Stuart Lewis from University of Edinburgh and published in ‘Research Data Management Technical Infrastructure: A Review of Options for Development at the University of Sheffield’. The workshop provided useful feedback on the architecture, including annotating the functions of various components and simplifying the repository. A revised version of the architecture incorporating these comments can be seen below.
Draft Top Level RDM Architecture for Feedback
The architecture follows the ideal data and metadata trails from the researcher into various institutional research systems and out into national and international data centres and the international scholarly communications systems.
The architecture chimed with many of the participants and it was seen to hold all of the components necessary to meet with the EPSRC’s policy mandates on research data and a good base to make decisions on developing an internal RDM infrastructure or shared service. This architecture is still draft and open to consultation, so please comment or contact John Kaye at Jisc if you have any feedback on the architecture. Jisc will also look to incorporate the findings of the Draft Research Data Management (RDM) Infrastructure Components work led by DCC
At the workshop a number of institutions discussed how their institutions systems compared with the architecture and a full report on the discussions and other workshop outputs will be available in early August. We heard from two different universities about their institutional scenarios and experiences in more detail. Loughborough university presented their RDM system based on Figshare, Symplectic, Arkivum and DSpace. King’s College presented a contrasting prototype system called crypta that is based on SharePoint as part of the Office365 package and as such benefits from free EEA storage.
At the end of the day Jisc presented it’s first thoughts about what a conceptual ‘all in one’ RDM service might look like. This type of service may be of interest to ‘greenfield sites’ with little or no existing RDM infrastructure and limited capacity or resources to develop or procure their own services. Within this presentation came the need to define the ‘stack’ behind this architecture. The image below is a simplified, layered version of these initial thoughts and is again open to comment and feedback. It focuses on simple user friendly user interfaces for the researcher to interact with; a middle layer of repository software, data registries, API’s and archival management; a storage layer with active, access and archival storage and a preservation layer, which Jisc is currently gathering requirements for.
Draft Simplified Hardware/Software Stack – Work in Progress for Feedback
Jisc would like to thank the participants at the workshop for their valuable input and welcomes comment from the community on the draft concepts outlined in this post. A full write up of the day will be available on this blog in early August and resources from the day can be found on the event padlet. A larger community event around Jisc RDM shared services is planned for early October 2015.
About the author:
John Kaye is a senior co-design manager within the research team at Jisc. He is also responsible for the Research at Risk work around shared services for research data. You can follow him on twitter @JohnPKaye.