As data becomes an even more critical part of daily operations at institutions and large universities supporting research projects, leading organizations often centralize their data management and storage systems to promote standardization, improve access and benefit from economies of scale. This university data center supports the research efforts of over 50 different campus groups. They offer various service level agreements (SLAs) based on the performance of the storage. Each research group is billed for the amount of storage they use based on the SLA they select.
“Our managed data can be human, application or machine generated, and it has been accumulating for years. StorCycle helps us provide the research groups we serve with a better way to manage large amounts of project data long-term. We can implement rule-based file movement across all storage tiers and easily archive, search and restore all data associated with a project.”
– Storage system administrator at university IT department
The university has standardized on three storage performance levels: an SSD tier, a high-speed disk tier and a NAS tier targeted for archive of projects. While NAS is the lowest cost repository in the university’s storage architecture, archiving large amounts of fixed content on NAS becomes extremely expensive over time. Researchers have asked for a lower cost solution for archiving.
The university actively encourages the researchers to move data off of the high-speed Primary Tier with a bill-back system that offsets their costs. The researchers have access to the data, but they have no way to identify what is actively used, what needs to be archived or what data is orphaned. Furthermore, when the university runs out of high-speed storage, they don’t always have the funds to acquire more before the offsets come in, which can hamper research efforts. Finally, grants for research often require storage of the data for periods longer than the researchers work for the lab or university involved.
The university chose to upgrade their storage environment with Spectra’s StorCycle storage management software and a tape library. Their cost for automated tape will be around 2.5¢/GB and additional copies on tape will be around 1.2¢/GB, relieving pressure from the most expensive Primary Tier storage that they are responsible for purchasing and maintaining. StorCycle offers a seamless view across all of the storage it manages, giving both the university and researchers access to all data across the Perpetual Tier. Researchers can use StorCycle’s scanning capabilities to identify and target inactive data sets for archive.
- Seamless access across all tiers of storage
- Project Archive feature
- Search capabilities by project name or other tagged metadata
- Project archive manifest tracks historical data movement