There are a variety of options for storing large-scale research data at UVA. Public and internal use data storage systems can be accessed from the Rivanna high performance computing system. Sensitive and highly sensitive data can be stored and accessed within the Ivy secure computing environment. Information Security at UVA provides an overview of the data sensitivity classifications.
UVA graduate and undergraduate students are not permitted to request storage accounts. This must be done by their faculty advisor[s].
2024-02-26 4:15 pm EST:
The update of Reseach Computing’s ticketing system instance has been completed and the history of all user tickets has been preserved.
On 26 February at 9:00 am EST, RC engineers will switch to an alternate data transfer mechanism between the legacy Research Project Storage filesystem and the new Project Storage system. The old and new Project storage filesystems have to be unmounted temporarily on all Rivanna nodes and the UVA Standard Security Storage DTN at 4:15 pm EST to complete that change. Project storage is expected to become available again at 6:15 pm EST. After the switch users will no longer have direct access to the legacy system. Files will be staged to an intermediate location for users to copy. To facilitate the migration process, a text file with a complete list of your old Project storage files has been placed into a new /stagedproject folder on Rivanna. You may use that list to indicate which folders or files should be prioritized for transfer using our data transfer request form. Additional information about the file transfer efforts and Project Storage system status is available on our Data Migration status page.
Public & Internal Use Data Storage
Public data are intentionally made available to the public. Examples of public data in research computing include:
- Data intended for a public website
- Datasets downloaded obtained from a publicly available source
- Open source code
Internal use data are classified as public records available to anyone in accordance with the Virginia Freedom of Information Act (FOIA) but are not intentionally made public. Examples of internal use data within a research computing context include but are not limited to:
- Files collected or generated from experiments that contain no personally identifying information (PII) or otherwise protected information
- Models, scripts, and logfiles
- Preliminary analyses
|Research Project Storage
|Research Standard Storage
daily snapshots for 1 week
|Rivanna, mountable on local workstation
|Rivanna, mountable on local workstation
|Ideal for long-term storage of data to be accessed from Rivanna. Research Project storage is ideal for sharing data within a research group and for running jobs with smaller files.
|Budget solution for storing data that can be accessed by a personal computer or Rivanna. It is not recommended to run Slurm jobs against research standard storage unless absolutely necessary. File operations on Research Standard storage are slower than on Rivanna
Research Project storage.
1Snapshot files are uneditable backup copies of all the files and folders in your account, taken at a daily interval. The system keeps these snapshots for a week. Snapshot files are deleted sequentially after a week has passed. This saving method is useful for human error prevention as any accidentally deleted files may be recovered. Look to our FAQ page to learn how to access your snapshots.
2Replication is a data management process that stores copies of data fragments over a distributed cluster or database. By having replicated data across each node or server on a given database, data can be accessed more reliably than data that only resides on a single server. This saving method is useful for disaster scenarios where if data is stored on multiple disks, and one disk fails, the data is still accessible.
3Backup files are copies of files that are stored on a separate disk storage than that of the original copies. Files may be backed up on a separate disk storage or within cloud storage. Backed up files are not synced with their original, so any edits to the original are not reflected on the backup. This saving method is useful for disaster scenarios where if the original disk storage is unsavable, the backups may still be accessible.
Sensitive & Highly Sensitive Data Storage
Highly sensitive data are data that require restrictions on access under the law or that may be protected from release in accordance with applicable law or regulation. Highly sensitive data (HSD) currently include personal information that can lead to identity theft. HSD also include health information that reveals an individual’s health condition and/or medical history. Examples of HSD include, but are not limited to:
- Personally identifying information (PII) such as social security number, passport number, driver’s license number, military identification number
- Information defined by the Health Insurance Portability and Accountability Act (HIPAA) as protected health information (PHI)
- Cardholder data (CHD) such as credit card numbers, primary cardholder account numbers, cardholder names, expiration date, and/or service code
Sensitive data is the default classification for all data that is not explicitly defined as highly sensitive data, may be held from release under FOIA, or that is not intended to be made publicly available. Examples of sensitive data include:
- University ID numbers
- FERPA-protected student information not covered by the definition of highly sensitive data
- Personnel and financial information not covered by the definition of highly sensitive data, but not intended to be public
- Any information that doesn’t fit into the other three categories
|Ivy Central Storage (ICS)
|Ivy Virtual Machine
|Ideal for long-term storage of highly sensitive data and is suitable for computation with smaller file sizes. Files stored in ICS are read-write only.
Researchers who request space on ICS must first request an Ivy account using the Ivy request form. Further information on Ivy and the ICS can be found here.
Ivy Central Storage is accessible by using Globus and connecting to the Ivy DTN.
Storage requests can be made via this form: