Research Data Architectures in Research Institutions IG Charter

Revised version (26/6/2018)

Introduction

The Research Data Architectures in Research Institutions Interest Group is primarily concerned with technical architectures for managing research data within universities and other multi-disciplinary research institutions. It provides insight into the approaches being taken to the development and operation of such architectures and their success or otherwise in enabling good practice.

An institution’s research data management infrastructure consists of more than just a data repository and discovery mechanisms. It includes the underlying storage technologies, the networking, hardware, system interfaces, authentication mechanisms, data brokers, monitoring platforms, semantic interoperability tools, long-term preservation services, high-performance and high-throughput computing facilities, data science platforms, and potentially many other technologies that process data and control the flows of data and metadata between systems.

Seamless data interoperability and movement between different systems both local and in national or disciplinary services is a particular challenge at present, given the need to provide researchers with a smooth and efficient user experience – a key requirement for any research data service. Governance and policies, project management environments, and communications platforms are also vital elements in shaping and informing IT architectures, as is the management of business information associated with research.

This IG seeks to understand the various architectures used by institutions globally, identify pain points within those architectures, and learn from those who have overcome or avoided those pain points.
The general approach of this IG is to encourage discussion about architectures and enable interested parties to collaborate and learn from one another. Many institutions are at present planning and working towards overarching data management architectures, and there is a legitimate concern that without such a forum as is provided by this IG institutions will relive the same experiences and repeat the same mistakes as their peers.

User scenario(s) or use case(s) the IG wishes to address

The Research Data Architectures Interest Group treats the following user scenarios:

• Knowledge support for research data infrastructure architects, and project and service managers
• Knowledge sharing regarding enterprise architecture best practises  at multi-disciplinary research institutions
• Development of a greater understanding of objectives and goals between different stakeholders at research institutions, including  management, infrastructure developers, research data engineers,  IT systems architects, and technology vendors.

Objectives

The main themes of the IG are

• Exploring how diverse tools, technologies, and services can be integrated to meet the evolving needs of researchers in research institutions.
• Considering interoperability between institutional research data infrastructures and (inter)national or discipline-based infrastructures
• Understanding the different institutional approaches to governance structures and business processes in responding to research ICT demands (e.g. capacity planning/forecasting for storage)
• Sharing case studies of solutions developed by data infrastructure projects in research institutions
• Presenting technical innovations and ideas that can further the development of integrated research data infrastructures
• Agreeing best practice relating to research data architectures in research institutions

This group has connections with some other RDA groups, such as:

• National Data Services IG – research institutions frequently support, use or deliver services and solutions provided by national data service vendors or operators, and these therefore form part of an institution’s architecture
• Repository Platforms for Research Data IG and Research Data Repository Interoperability WG – repositories are an essential component of research data architectures, but need to be considered as one element of a larger infrastructure
• Data Fabric IG – technology and interoperability solutions for research institutions are an essential part of the bigger picture)Storage Service Definitions WG (formerly “WG QoS-DataLC Definitions”) – storage definitions, vocabularies etc. are important for achieving mutual understanding of research data architectures

The IG differs from these RDA IG’s or WG’s in following areas:

• The scope of the group is institutional-level solutions and architectures rather than for national or disciplinary-specific architectures, although it will be important to consider the relationships between these different levels
• The Group will discuss technological solutions at research institutions at the enterprise architecture level, not focus on a single element in isolation

Participation

The target audience of this IG includes anyone involved in research data infrastructure planning projects or services as well as researchers with an interest in systems, technologies and data flows at the institutions level. This includes research data infrastructure project managers, ICT architects, senior managers with responsibility for research IT services, developers, data managers, data engineers, and data scientists. Also representatives of the service or technology vendors and data industry are welcome.

Outcomes

• Outcomes of the IG:
• Shared knowledge about research data tools, infrastructures and architectures
• Knowledge base for best practises and lessons learned log
• More specific outputs will be determined by the participants in this IG, but might conceivably include:
• Shared repository of architectural diagrams
• List of technologies used for specific purposes, and their interfaces
• Landscape report and gap analysis

Mechanism

The Interest Group will hold regular meetings at RDA plenaries (if accepted in the schedule). Between Plenaries the IG will collect case studies and examples of good practice and add these to the open knowledge base. Group will have 1 - 2 web meetings between plenaries (possibly more, if it is needed).

Timeline

P10: BoF and start of the review process
P11: Starting the IG; Discussion about tools, best practices, use cases and knowledge base
P12: Templates for the best practices reports
P13: First collection of the knowledge base