From Bioinformatics.Org Wiki

Broadening our backup policy - ideas

• More offsite backups in different locations
  • cost of hard drive plus network drive enclosure
  • protects against loss of data due to any physical damage
  • very effective, except makes backup system harder to maintain - must be used in proper balance with other methods

• Raiding offsite backups
  • cost of hard drives
  • protects offsite backups from faults
  • cheaper than more offsite backups, but less effective at protecting from physical damage

• More frequent backups
  • decreases the size of loss from any faults (i.e. hard drive faults)
  • will require larger hard drives - or - will require deleting backups from the storage device much earlier so as to not overfill the backup drive
    • this may not be an issue depending on the average size of backup, and if we use techniques like incremental backups to help the process.
  • can be implemented with little to no change to current hardware

• Adding more drive to onsite RAID/time machine configuration
  • cost of hard drives
  • protects against immediate loss of work due to physical hard drive failure
  • time machine protects against loss of data due to any accident (so long as the hard drive with the time machine itself is intact)

• Offsite time machine
  • limits installation to CS network (at least until cat-6 cabling is installed)
  • Mac proprietary system is very effective at preventing against data loss
  • reasonably cost effective
  • similar to more offsite backups

Questions about offsite backup

1. What level of data integrity is appropriate/necessary?
2. Do we have access to any other locations for offsite backup? (current locations are E2-537 and the Plant Science building)
3. Can we integrate our storage or backup system somehow with IST?
   • NOTE: we are already storing our external drive in Dave Trebble's office
4. Are there any services on or off-campus that will provide additional data integrity?