The Research Data Lifecycle at UCR

UCR Research Computing provides a tiered storage ecosystem designed to support the entire lifecycle of your research data. Rather than choosing a single platform, we recommend a Lifecycle Strategy where data moves between tiers based on its current utility and access frequency.

🚀 Tier 1: High-Performance Analysis (Hot)

Best for: Active computation, temporary scratch space, and data currently being processed by the HPCC.

HPCC – GPFS (Parallel File System)

• Performance: Ultra-high throughput, optimized for massive parallel jobs.
• HPCC Rates and more info
• Note: This tier is for “working data.” Once analysis is complete, data should be moved to Tier 2 (Project) or Tier 4 (Archive).

🏢 Tier 2: Active Project Storage (Warm)

Best for: Lab shares, sequencing libraries, instrument data capture, and data that needs to remain “online” for frequent access but isn’t actively computing.

CephRDS (Secure Research Storage) — [PILOT PHASE]

CephRDS is our next-generation, on-premise storage platform designed specifically as the backbone for research project data.

• Status: Currently in Pilot. Ideal for CAREER grants and labs needing high-capacity repositories.
• Scalability: Starting at 2 PB and designed to scale seamlessly.
• Security: Built on NVMe-backed storage with 8k + 4m erasure coding and robust encryption.
• Accessibility: Supports NFS and S3; integrates directly with the HPCC.

SDSC Qumulo – Universal Scale Storage

• Use Case: Suited for everyday research needs, similar to an office NAS.
• Cost: ~$70/TB/year (Normalized). Optimized for on-campus use.

🤝 Tier 3: Collaboration & Documentation

Best for: Manuscripts, administrative files, protocols, and sharing lightweight files with collaborators.

Google Workspace (Google Drive)

• Included: 100 GB for Faculty, Staff, and Grads.
• Best Practice: Use for documents and small-scale data sharing. Not recommended for raw genomic sequencing or multi-TB binary data.

❄️ Tier 4: Long-Term Archive & Compliance (Cold)

Best for: Raw data retention, grant compliance (e.g., NSF/NIH 10-year rules), and disaster recovery.

Ursa Major Cloud Storage (GCP Archive)

• Branding: Part of the Ursa Major Cloud ecosystem.
• Cost: ~$15/TB/year (Our lowest cost per GB).
• Use Case: “Write once, read never.” Perfect for long-term preservation of raw reads or project snapshots.

AWS S3 Glacier Deep Archive

• Cost: ~$12/TB/year.
• Use Case: Alternative for extreme long-term preservation where 12-hour retrieval times are acceptable.

💡 Which Tier is Right for You?

If you are unsure how to architect your lab’s data flow, please schedule a consultation with our team. We can help you automate the movement of data between these tiers to minimize costs while ensuring maximum data integrity and security.