1. The HFS service
The HFS is an Oxford University centrally funded service providing data backup and long-term archive services to senior members, postgraduates and staff. The name HFS was originally an abbreviation for Hierarchical File Server. However that is no longer an accurate description of the service and HFS should now be considered as simply a name rather than an abbreviation of anything.
The HFS makes use of IBM's Spectrum Protect software (previously known as Tivoli Storage Manager, or TSM) running on the AIX operating system on several IBM Power servers to provide more than a dozen virtualised Spectrum Protect servers.
The HFS is widely and extensively used throughout Oxford. In the 7 days up to 19 September 2017, 4,438 clients accessed the backup and archive servers in 25,377 sessions, sending 118.7 million files and 141.8 TB of data. In total the backup and archive servers currently hold 3.842 billion files and 3083.6 TB of data.
We welcome any comments or suggestions about the service. Please contact us at firstname.lastname@example.org.
2. How it works
Your data is sent from your computer, across the university network to the HFS servers and is then ultimately stored on magnetic tape in an automated (robotic) tape library. Our servers and tape library are situated in a climate-controlled, secure location. Three copies of your data are made, each to separate tapes: one copy is held in the automated tape library while two copies are stored in separately located fire-proof safes. Access to the data is private to the owner and is normally available 24 hours a day, seven days a week.
The HFS employs two TS3500 IBM tape libraries dual-sited across two data centres. The online tape library offers 26 latest generation 3592 drives and currently offers an operational capacity of around 10PB which is, as the name suggests, online and available 24/7. Expansion frames and roadmapped advances in tape technology should allow this to reach 30-40PB over the next decade. The offsite tape library writes tapes for offsite copy and so has no need for extended storage and utilises 12 3592 drives of the same type as the onsite library.
Statistics on data intake etc can be seen on our statistics page.