Closure Report
Project Summary
Across CMVM there is a significant variation in the adoption of centrally supported storage services. Current storage solutions are understood to include University services such as DataStore, DataSync and OneDrive as well as unsupported and third-party storage technologies. Some of these may be putting data at risk of loss and potentially breach, and may not be backed-up, which is contrary to best practice and guidance.
It is not clear what the full range of storage solutions currently in use is, and therefore the level of risk which is inherent to the researchers.
This project sought to:
- Identify at a high-level the range of storage solutions that are in use.
- Develop a strategy for sustainable data storage practice and service across CMVM. The output of the strategy will signpost what good practice looks like. It will document the recommended College approach for the use of storage services, and suggest how to move from the existing storage solutions to the recommended approach.
- The recommendations are intended to encourage best practice and inform people about what they need to do to achieve and maintain effective data storage practices.
- It will create a process to help move data from higher risk storage to the recommended approach.
- As part of the proposals for research grant applications these will be reviewed, and potentially amended, to include the provision for information managements, data planning and storage requirements.
- Appropriate governance bodies will be identified to review, approve and recommend the strategies.
Objectives & Deliverables
|
O1 Storage Service Strategy |
Priority |
Outcome |
|
O1.D1. Document Storage Strategy for the College. |
M |
CMVM Data Storage Strategy (Restricted access) |
|
O1.D2. Document best practice guidelines for Research Data Storage. |
M |
Guidelines and support for researchers provided through the Research Data Service |
|
O2 Data Storage Policy Management |
||
|
O2. D1. Collate any College data management policies/practices supporting Storage Strategy. |
M |
Existing University level groups - policies and guidance existing in this area include:
Related regulations and policies that support this area include: |
|
O2.D2. Identify the appropriate governance group to approve Data Management Policies. |
M |
University level policies are approved by the Research Data Management Steering Group, which reports to the Digital Research Services Steering Group. This is University-level. |
|
O3 Storage Migration Service |
||
|
O3.D1 Identify Research Groups within CMVM who are accountable for storing research data. |
M |
All CMVM Research Institutes and Centres hold research data. Every individual (both members of staff and students) using research data is responsible for following data privacy legislation; the Principal Investigators are accountable for use of the data for each research project. |
|
O3.D2. Build an on-boarding and move process with appropriate approval and authorisation workflow to support moving to the recommended solution. |
M |
|
|
O4 Documenting current College storage service options and indicative costs |
||
|
O4.D1 Map the storage options that are currently in use, and those available and recommended. |
C |
Summarised in a great one-page table by Research Data Services: Quick Guide to Research Data Storage |
|
O5 Grant Application Process |
||
|
O5.D1. Update and publish the amended grant application form to include data management, data planning and storage requirements supporting planning for appropriate Research Grant Funding for continued and affordable use of the recommended solution. |
S |
|
Analysis of Resource Usage:
Staff Usage Estimate: 20 days
Staff Usage Actual: 21 days
Staff Usage Variance: 5%
Explanation for variance
The increasing in budget was caused by an extended timeline for delivery casued by lack of availability from key business stakeholders, and more recently from a change of PM following the departure of the contractor PM. The programme manager and sponsor managed the closure.
- PICCL 2 - March 2019: Delay to delivery and closure - lack of stakeholder availability
- PICCL 3 - July 2019: Delay to closure - lack of stakeholder availability
- PICCL 4 - August 2019: Change of PM
- PICCL 5 - September 2019: Delay to closure - lack of stakeholder, PgM and Sponsor availability
- PICCL 6 - January 2020: Delay to closure - lack of PgM and Sponsor availability
- PICCL 7, PICCL 8 and PICCL 9: Delay to closure due to the Covid-19 outbreak - lack Sponsor availability
Outcome
Data volumes are growing fast and data management complexity is increasing. Data and capacity mitigation strategies are required for effective and efficient storage utilisation and risk reduction.
To develop an effective storage strategy, mitigating the rise of both data growth and data complexity started with data policies and governance. Aligning technology decision making with governance objectives and workload requirements.
Data has variable value to the organisation and it needs to be differentiated. Some bits are more important than others and some needs to be accessed faster.
Data differentiation, through data governance, has been addressed. Without differentiation of the data on the basis of business value and criticality, we would not be able to make technology differentiation e.g. storage tiering and data protection.
This strategy is about the data, understanding what is being supported and why. Deciding what and how data is stored before deciding where. The needs of our workloads and the governance requirements of our business drove our storage decisions and the technologies we have adopted.
Current and future capacity needs for current and future data drivers were identified. Evaluating current storage ability to meet those needs helped us to discover any necessary additions to meet the requirements.
Governance requirements and constraints that exist across the organisation that are specific to our workloads were identified. This identification, which involved our wider business, has a tremendous benefit for managing data growth.
These data and capacity mitigation strategies help effective and efficient storage utilisation and risk reduction. The strategy explored data and capacity mitigation strategies to maximize the effective utilisation of our storage offerings.
Key Learning Points
Many funders, in particular charity funders, have not fully considered in grant funding the very real costs associated with services to store (and analyse) research data. This has led many researchers in CMVM to develop a wide-range of solutions to store data, and, for those that do not use the recommended University solutions, there is a significant risk of data loss through lack of back-up and restore facilities, and local support requirements to run local solutions.
This strategy will likely take 3-5 years to implement, as data management needs to be planned into new grant funding projects.
Outstanding Issues
MVM had expected to select a research group and provide funding to support and document the migration process from existing unsuported storage (i.e. Dropbox) to the recommended storage (i.e. datastore), as an examplar and to provide SOPs and guidance for other to follow. The sponsor has not had the capacity or funding to allow this following the Covid outbreak. The Programme and Portfolio owner have agreed to do this owrk as part of BAU (see PICCL10 -Agreement to close project with outstanding actions). A risk on non realisation of teh benefits of this project has been recorded at the Programme level to keep track of this item, see PRICCL 4 - Benefit realisation for MVM311 CMVM Data Storage Strategy