The UT-StorR Archive and Archiving process

UT-StorR is a long term archival storage system intended for the storage and long term preservation of large collections of research data. UT-StorR is made up of an 800 TB Lustre high performance parallel file system with a Tfinity T950 tape robot currently capable of storing 6.3 petabytes of data on tape. The integration between Lustre and the tape system is provided by an HPE Data Management Facility (DMF) hardware and software capability. The migration of data from Lustre to tape is fully automated via DMF policies. UT-StorR will migrate off to tape any file that is idle for more than 24 hours.

It is important to remember that UT-StorR is an archive, not a backup system or an HSM (Heirarchical Storage Manager). A long term archival storage system is a system designed for preservation of data that is no longer actively used and is a place to store results data no longer actively used, for primary data, such as, initial datasets from instruments (Spectrometers, DNA sequencers, CT scans, MRI scans, etc.), and for preservation of research data collections over long periods of time as would be described in a research proposal data management plan. A key phrase related to long term archival storage would be “write once, read seldom”. A backup system involves frequent incremental copies of data or file systems and occasional full dumps of data or file systems. An HSM involves multiple “tiers” of storage, from small, fast and expensive (NVME, SSD) to big, slower, and cheaper (HDD, Tape), with automatic migration between tiers as the file ages. UT-StorR is not designed as a backup system or as an HSM. UT-StorR is designed as a long term storage platform for research data for the University of Tennessee.

Examples

An example of appropriate use: A project or sub-project task has wrapped up, the data and artifacts used for that task/deliverable are not needed for future use but may be needed later or need to be preserved in accordance with the projects Data Management Plan. The associated data and artifacts are moved to the archive and curated with metadata descriptors. 

An example of inappropriate use: PI wants a backup snapshot of a project directory. PI copies project directory to archive several times a month with minor changes in the directory data with no curation. 

Use Cases

The intended Use Cases for the UT-StorR long term archival storage system includes: (1) researchers who use the ISAAC campus clusters and have large amounts of primary and results research data and data collections that are no longer actively used or used very infrequently, (2) UTK Core Facilities that manage instrument and other data and data collections for a community, and (3) in cooperation and coordination with UTK Library, the data storage by researchers, research projects, or Core Facilities to meet the data management plans of funded research projects.

Access to UT-StorR

If your storage needs meet any of the Use Cases mentioned previously request access to UT-StorR by filling out our service request form here. HPSC staff will get back with you to discuss your long term archival storage need, provide training and UT-StorR overview information, and get you set up for access to UT-StorR resources and services. Access to UT-StorR requires mandatory training by any researcher(s) or research teams that need to use UT-StorR for any of the three Use Cases.

Upon processing the UT-StorR access request, mandatory training will be scheduled and then a directory will be created in the appropriate policy directory of the UT-StorR file system (the /lustre/utstorr landing pad). This directory will be named with an appropriate identifier (e.g. UTK0022) and the PI’s group will be granted permission to the directory. Mandatory training and an existing ISAAC project are required before access is granted to the archive.

Login

The UT-StorR landing pad is accessible via the dtn2.isaac.utk.edu data transfer node. The file system is /lustre/utstorr.