Many SANs grow beyond their original expectations, without due care and structured planning. New switches are added to serve increased data capacities and host connections. The mesh of switches and fibre devices can feel like a web, with the SAN administrator trapped in the middle.
It is at this point that many skilled SAN administrators become skilled fire-fighters. There is no greater shame than when an experienced SAN admin spends more hours trying to deal with compounded issues rather than planning the technologies and processes to take their storage environment to the next stage of delivery capabilities.
And so the cycle continues. With each problem experienced, quick fixes are needed, causing increased complexity and ultimately leading to more time spent managing the creaking build.
Whilst no one can predict the future, three basic rules can make your mid-term and longer-term goals for a SAN easier to achieve.
1. Create a workable, flexible physical environment for your SAN. As your business grows, so does the impact on your data centre. Taking more time when racking and cabling switches will minimise time allocated towards sorting out issues with tens / hundreds / thousands of erroneous stray cables stretching across switch cabinets and hanging tightly from switch ports. Get it right early, and potentially thousands of pounds can be saved in the time and effort used to run an operating environment
. 2. Create a workable, flexible logical environment for your SAN. Once again, as the number of Fibre Channel devices in your environment grows, so does the management overhead to administer your storage network. Think not only of the connections that you add to your SAN, but also what you might need to do with them in the future. Deliver present-day needs whilst future-proofing what tomorrow might bring by adhering to industry best practise. Simple things like implementing as few zone members in zone sets as possible can ease such tasks as future system migration by allowing major fabric configuration changes to occur with controlled impact.
3. Document everything. Use a Systems Operations wide CMDB. If you do not have one of these, take some time to develop an easy-to-update and -read centrally managed documentation library, and help your existing team (and those future administrators yet to join you) quickly understand the environment they are dealing with. There is nothing more painful than undertaking a retrospective audit of a systems environment that has outgrown the expectations of those who built it.
Everyone knows that building a house without a strong foundation will lead to problems -- and the bigger the house gets, the greater the potential risk of it crumbling around you. It's the same with a SAN. In a SAN, the simple things are often the most effective, and as SANs get more complex, the basics become even more important.
About the author:Allaster Finke is a senior consultant for Glasshouse Technologies (UK), a global provider of IT infrastructure services. He has more than seven years experience in the design and delivery of IT solutions, with a particular focus on SAN, storage and backup technologies.