* How to delay buying more storage hardware
Why should you continue to over-provision your storage assets? Even today, many companies are still without some of the most basic capabilities when it comes to storage management. Lack of storage resource management tools means they don’t know what assets they have, or how efficiently they are being used. Many organizations also lack a useful planning process, so they have little idea of what they will need in order to service their clients during the coming years.
As a result, many companies of all sizes still go through the quarterly knee-jerk exercise of buying increasing amounts of disk storage hardware as a hedge against unknown future demand. These preventive measures are expensive, and often not even necessary.
Taking stranded storage and moving it into a manageable environment – onto a Fibre Channel or iSCSI storage-area network (SAN), for example – is an easy way for companies to save money.
In today’s world, storage is often used at a rate of between 30% and 60%, with the rest going to waste. Furthermore, storage that is widely dispersed across multiple SANs, or that is attached directly to a series of servers that cannot be centrally managed, is both difficult to maintain and expensive to manage. The operational expenses are likely to be killers.
If your application servers have lots of direct-attached storage (DAS), you may well be sitting in the middle of a mother lode of unused storage that is ready to be exploited. If, for example, you have 500T-bytes of DAS or remote storage at your site, it may well be that half that total is going unused without anybody paying attention to it. What would be the value to your company if you didn’t have to manage another 250T-bytes of storage next year? Spending a few dollars today on learning how to efficiently manage existing assets can often result in delaying by as much as a year a large planned investment in additional disk drives and arrays.
If you are going to reclaim your site’s existing unused capacity, the first step must be to identify existing storage that is under-used or cannot be efficiently managed.
Unfortunately, much stranded storage is also DAS – and DAS, which was never meant to be centrally managed, rarely complies with industry standards. Because of this, the auto-discovery techniques that you may be using to identify resources on SANs or network-attached storage (NAS) devices are likely to be of limited use. In any case, there is likely to be little opportunity for such tools to determine any pathways between applications and storage assets attached to other application servers.
One nice way around this is to bite the bullet (small caliber) and have an outside organization come in and do a storage assessment for you.
Outsourcing a job like this can make a lot of sense if the firm you contract with already has a tool set that can be applied to your situation. If you can actually find such a company – I know of a few – satisfy yourself that it is vendor-neutral and thus has no axe to grind when it makes its assessments. When you engage with the company you may find that in short order you have identified assets that can be used in place of the ones you were already planning to buy.
The best feature of all this is that the relatively small expense involved in engaging with such a firm may result in almost immediate payback.
