Where’s the Data? Love Data Week and the Journey Home for Our Research

Guest post by Nicholas Owen DPhil, UCL Advanced Research Computing

It’s a question that every researcher has asked, usually with a rising sense of panic at 4:00pm on a Friday – “Where is the data, which version is it?”

Is it on the USB drive labelled “Backup 2023”? Is it sitting on the department’s or group’s NAS box that has been humming ominously under a desk since the previous grant cycle? Or is it trapped in a siloed departmental server that requires three different VPNs to access?

For Love Data Week this year, the theme is “Where’s the Data?”—a question typically focused on the joy of discovery and re-use. But for those of us working as Data Stewards, this question is literal. We are driving a major initiative to bring our research data officially under UCL management, shifting it from the periphery to the core of our infrastructure.

This is about more than just storage; it is about stewardship. By centralizing these assets, we are working to benefit both individual researchers and the university as a whole – securing our institutional knowledge to ensure maximum visibility, impact, and potential for reuse, whilst providing best practices on data management.

For too long, the answer to “Where’s the Data?” has been “everywhere and nowhere.” As part of Love Data Week, we want to talk about our work at Advanced Research Computing (ARC) to change that and how you can get involved. We are moving away from the “Disk Drive Drawer of Doom” and toward a centralized, resilient platform that doesn’t just store data, but aims to seamlessly connect to compute, archiving, and publishing services that bring research to life.

The Era of the “Digital Shoebox”

Historically, data storage was treated as a personal problem. You generated data, bought a hard drive, and stuck it on a shelf. But as our datasets have grown—from gigabytes to petabytes—this model has fractured.

We also see a drift toward “free” commercial cloud solutions that offer short-term convenience but hide long-term risks. As these services are often tethered to personal accounts rather than institutional ones, they create a precarious dependency: when staff leave the university, the data often leaves with them or remains locked behind an inaccessible login. What looks like a quick fix effectively becomes a governance black hole, putting long-term access at serious risk.

Beyond the risk of data loss, there is a more immediate threat: cybersecurity. That NAS box humming under a desk or the server in the corner often falls into the category of ‘set and forget.’ As months turn into years, these devices frequently go unpatched and unmanaged, becoming invisible to central IT but highly visible to attackers. In today’s landscape, an outdated, unpatched server is not just a storage device; it is a potential entry point for ransomware that can lock up not only your research data but potentially compromise the wider university network. Keeping data on unmanaged hardware is effectively leaving the keys in the door.

Whether it’s trapped in a personal cloud account or on a multitude of physical drives, we see valuable data vulnerable to hardware failure, loss, or simple obsolescence. More critically, we see silos. When data lives in a departmental NAS box, it is cut off from the wider UCL ecosystem. It is difficult to share, expensive to maintain individually, and often impossible to audit for compliance.

To truly “Love Data,” we must stop treating it like a static object to be hoarded and start treating it like a dynamic asset to be managed.

Building the Central Hub: Research Data Platform

Our initiative is not just about buying a bigger hard drive for the university. It is about building a platform that adheres to three core principles:

1. Resilience

Research data is the currency of our institution. Losing it is not an option. By centralizing onto a professional, enterprise-grade platform, we move away from the “single point of failure” model (i.e., the drives under your desk). We are implementing robust redundancy, automated backups, and disaster recovery protocols that no single research group could afford or manage on their own. When you ask, “Where’s the data?”, the answer is: Safe.

2. Cost-Effectiveness

Storage at scale is cheaper than storage at retail. By consolidating our purchasing power and management, we want to offer tiers of storage that fit the data lifecycle. We can provide high-performance flash storage for active analysis and will soon offer low-cost object storage for cold archiving, all without the researcher needing to negotiate with vendors or manage hardware refresh cycles. We remove the hidden costs of shadow IT—the power, cooling, and administrative time spent keeping old servers alive.

3. Utility

A vault is useless if you can’t get inside. The goal isn’t to lock data away; it’s to make it usable. The research data platform is being developed to be accessible, indexable, and searchable. It transforms data from a “file on a disk” into a discoverable resource that can be easily retrieved by the people who need it, when they need it.

From Policy to Partnership: How We Work with You

We know that “centralization” can sound like loss of control to a researcher. That is why our approach is built on partnership, not just policy.

Our Data Stewards don’t just send out a compliance checklist; we work directly with individual departments to understand their unique data needs. Whether that’s millions of tiny sequencing files in the life sciences or massive 3D scans in the humanities. We sit down with your team to audit legacy data on dusty drives, map out current workflows, and design a migration plan that minimizes disruption. We are not just offering a destination for your data; we are acting as the removal team, the architects, and the security consultants to ensure that when your data moves, your research accelerates.

The “Seamless” Experience

Perhaps the most exciting part of this centralization is not the storage itself, but the ecosystem connecting to it. In the old model, moving data from storage to a high-performance compute (HPC) cluster was a manual, often painful bottleneck. We are dissolving the boundaries between storage, analysis and publication.

• Integrated Compute: Imagine your data residing on the platform and being instantly accessible to our HPC clusters. No more file transfer protocols that take days from external drives/NAS boxes. You run your analysis where the data lives.
• Archiving: When the project is done, the transition from “hot” active storage to “cold” archive storage shouldn’t require a migration plan. It should be a policy setting. We aim to automate the lifecycle so that data ages gracefully without vanishing.
• Publishing Services: A major goal of research is dissemination. By centralizing the data, we can build direct pipelines to data repositories and publishing services, effectively bridging the gap between “working data” and “published data.” Through the Research Data Storage Service (RDSS) and the UCL Research Data Repository (RDR), we ensure data lands in the appropriate service. We can streamline the issuing of DOIs, ensure metadata compliance, and meet funder mandates for Open Access—all without the researcher having to manually re-upload terabytes of data to external systems. Crucially, we recognize that one size does not fit all. We actively promote depositing data in the most domain-appropriate resource for your field. In these cases, the RDR acts as the central signpost. By using metadata-only entries, the RDR creates a permanent record that points to the external location. This ensures that even when data lives in a specialized domain repository, the university maintains a clear answer to the question “Where is the data?”

Loving Your Data Means Giving it a Home

“Where’s the Data?” shouldn’t be a scary question. It should be the start of a new discovery.

By moving away from the fragmented landscape of external drives and siloed servers, we are building a home for our university’s research that is worthy of the effort that went into creating it. We are building a system that is resilient enough to trust, cost-effective enough to sustain, and useful enough to power the next generation of breakthroughs.

So, this Love Data Week, let’s celebrate not just the data itself, but the infrastructure and teams that keep it safe, accessible, and ready for the future.

Ready to Bring Your Data Home?

The transition from external drives to a centralized ecosystem starts with a single step. The Research Data Stewards team at ARC is here to guide you through it.

We know that moving terabytes of data can feel daunting, but you don’t have to do it alone. We are ready to help guide you migrate your active and archival data from external sources and siloed hardware onto the RDSS.

• Start the Move: If you are currently relying on USB drives or aging NAS boxes, contact us through MyServices to schedule a consultation on moving your data to the RDSS.
• Get Advice: Unsure about file organization, metadata, or compliance? Ask us about best practices for managing research data to ensure it remains useful for years to come.
• Reach Out: You don’t need a fully formed migration plan to talk to us. We are here to help you scope the work.

Start the conversation today. Contact our team at researchdata-support@ucl.ac.uk and let’s make sure your data is safe, accessible, and ready for whatever comes next.

The UCL Office for Open Science and Scholarship invites you to contribute to the open science and scholarship movement. Stay connected for updates, events, and opportunities.

Follow us on Bluesky, LinkedIn, and join our mailing list to be part of the conversation!

Share this post on Bluesky