Research Data Management

OVERVIEW

In general, it is the responsibility of the researcher to include proposal monies for computing resources, storage, backup, preservation, high performance computing or use of the Mines Institutional Repository. Below are backup and storage options available to Mines researchers.

ACTIVE PROJECT STORAGE PROVIDED BY IT&S

ADIT DRIVES

• Suitable for data-to-day office and student work
• Suitable for some research data needs, but storage is typically on the order of ~50 gigabytes (GB) per user
• Is generally accessible on the servers of Hornet and Talon (Windows file servers) and Fermat and Isengard (Linux file servers)
• Drives are backed up

OREBITS STORAGE SERVICE

• Suitable for most research data needs
• Accessible to those having a Mines Multipass ID
• Granular permissions
• Purchased in increments of a terabyte
• Customizable options based on need (single-copy, replicates, backup)

BACKUP SERVICES (FOR LAB COMPUTERS, LAPTOPS)

Currently, the best backup option for your research lab computers or laptops is to use built-in Mac and Windows software. Use external drives/disks designed for backups as opposed portable media like USB drives, DVDs or CDs. These external drives should be at least 50% larger than the data/disk you want to backup.

• Mac: use the built-in Time Machine. Create both a backup disk and offsite backup disk.
• Windows: use Windows backup. Save to your hard drive and another medium.
• UNIX/Linux: use rsync with an external USB disk
• Purchase of cloud computing services in accordance with the Mines Cloud Computing Policy (under development)

END-OF-PROJECT STORAGE (MINES REPOSITORY)

• Suitable for end-of-project deposit of materials in order to meet requirements of making data publicly accessible
• Suitable for small to medium sized datasets
• Space is limited; select the highest priority items
• Free < 10 gigabytes (GB)
• Creates backup copies

HIGH-PERFORMANCE COMPUTING (HPC)

• Suitable for active projects only; not available as permanent storage
• Either purchase your own node or use the supercomputer
• See HPC for more information