There’s lots of misconceptions about backup – here, I want to present 20 of them:
- Backup is an IT activity. It’s not. It’s a corporate insurance policy.
- Backup is a function of ILM. It’s not. It belongs in ILP, which is not the same thing. See “Think Backup Belongs in ILM? Think again“.
- Backup is not a production activity. It is. See “Backup is a production activity“.
- You don’t need a support contract for your backup environment. You do. See, “Perils of an Icarus support contract“.
- You can migrate hardware you’re phasing out of production into backup and DR. You can’t. See, “If you wouldn’t drink it, don’t cook with it“.
- Backup failures happen. They don’t. (Potential recovery failures happen.) See, “There is no such thing as a backup failure“.
- You can have a backup system without a zero error policy. You can’t. See “No Zero Error Policy? No backup system“.
- Backup stops at servers and the occasional desktop. It doesn’t. See “What don’t you backup?“, and “But where does the DPA fit in?“
- Deduplication can be tacked into a backup solution and immediately solve your capacity problems. It’s not that simple. See “7 common problems with deduplication“.
- Inclusive backup policies work. They don’t. See “First, backup everything“.
- Backup (and more generally, IT) issues are technology issues. They’re usually not. See, “Technology is rarely the issue“.
- Backup capacity issues are solved by adding more capacity first. You should add backup capacity last. See, “A basic data lifecycle“, and its follow up posts.
- Archive is backup. It isn’t. See “Archive is not backup“.
- A complex backup environment is a good backup environment. Wrong. See, “Of unicorns and horses“.
- Backups should be rigorously cost controlled, with every saving aggressively pursued. Wrong. See, “Backups are not about being miserly“.
- Cross-site backups allow you to avoid cloning or backup duplication. Nope, definitely not. See, “You can’t escape cloning with cross-site backups“.
- Backup sucks. Wrong. It’s actually one of the best jobs you could hope for. See, “Backup does not suck“.
- Tape is dead. No, the usage strategy has evolved, that’s all. See, “Direct to tape is dead, long live tape“.
- Capacity planning can be done quickly. No, it can’t (unless you have the right tools). See, “Capacity planning – quickly or properly?“
- The dense filesystem problem doesn’t really happen. Yes, it does. See, “In-lab review of the impact of dense filesystems“.
I see this article as a good compilation/index of your posts.