Upgrades seem to be governed by the IT version of Murphy's Law: If it's not necessarily true that what ever can go wrong, will go wrong, it's certainly true that if system A needs upgrading, then system B will require adapting as well.
A SearchStorage.com reader recently wrote to our storage management experts to ask for help determining whether his organization should concentrate its efforts on implementing a SAN solution, or on improving its existing backup software and hardware. He wrote: "My question is: Which one comes first? I believe they are integrated concerns. We need to pursue a backup solution (due to current issues), but do not want to over-purchase on the tape arrays and hardware. What are other organizations doing?"
In the short term, says Norbert Haag, "Whatever makes the biggest bang for the buck comes first."
In the long term, however, Haag recommends implementing a SAN first to increase the choices of backup system. He also advises users to take pressure off the backup by categorizing data. This minimizes the amount of storage required, by backing up only valuable data and data that has changed.
Determining which data is valuable data is not IT's domain, however. IT can only support the valuation process made by the owner/user of the data. Haag suggests implementing a "charge-back" policy that puts the cost for storage back with user.
Jamie Gruener comes at the problem from a different angle. He recommends beginning your assessment
- How much data will be migrated to the SAN?
- What is the average backup load per week, month and year? How is it changing?
- What is the corporate standard for storage system capacity? (And if there isn't one, is there an expectation that the storage arrays and tape systems will be fully utilized immediately, or is there a need to under-provision for either performance or availability reasons?)
Read Norbert Haag's answer to this question.
Read Jamie Gruener's answer to this question.
This was first published in June 2004