Secure, governed environment for sensitive and large-scale research datasets at UCLA.
Role: Platform lead and governance architect Focus: Secure data environments, P3/P4 data governance, high-throughput research computing
Overview
Before 2021, UCLA had no compliant environment for multi-terabyte restricted datasets. Researchers with sensitive data — health records, administrative data, proprietary datasets — had no institutional home for that work.
We licensed and implemented Redivis to fill that gap.
What we built
- A governed, compliant environment for P3 and P4 sensitivity data
- Workflows for Data Use Agreement (DUA) management and access control
- Onboarding and governance processes that scale without routing every request through senior staff
- Integration with campus identity management and audit infrastructure
How it works
Researchers apply for project workspaces. DUAs and access controls are enforced at the platform level. The governance model is designed so that routine requests are handled through standard workflows, reserving escalation for genuinely complex cases.
This is intentional: a platform that requires senior staff intervention for every transaction is not sustainable.
Outcomes
- Grew from 5 workflows and 444 GB processed to 2.9 PB processed in 2024
- Supports research published in APSR and PNAS
- Contributed to Carceral Ecologies work (318-minute pipeline reduced to 2 minutes) and associated $5M grant
- Supported BioCritical Studies Lab analysis of deaths in custody
Collaborators
- Redivis platform team
- UCLA Office of Research (compliance)
- UCLA IT (identity management, security review)