Monday, 31 August 2009

Adding New Storage Frugally

Even In Times of Tightening Belts, New Storage Systems Are Possible

With the economy on everyone’s mind, adding new storage to an existing data centre can be a challenging process. Strict budgets and a tendency to nix already green-lighted projects can make it difficult for IT to gain the dollars to move forward.

In such times, though, IT has a responsibility to make sure it is adding storage wisely and using what is already there in a cost-effective manner. And even though SMEs might not be able to afford the latest and greatest gear available today, there are smart ways to implement upgrades to gain maximum bang for the storage buck.

Only Add When Necessary

The cardinal rule in the modern climate is not to add storage that you don’t need. Obvious as it seems, it is one of the most commonly abused rules based upon an entrenched overbuilding mentality. As result, there is usually an abundance of unutilized storage sitting around the data centre.

“Companies need to ask if they know what their capacity utilization is,” says Jim DeCaires, storage product marketing manager at Fujitsu America. “They also need to understand how effectively they are using their storage and if they can capture unused capacity. Most storage capacity is underutilized.”

Before requesting the purchase of additional storage, IT should achieve an understanding of current utilization. It can be very embarrassing - and financially damaging - if management finds out it just spent $50,000 on storage hardware it didn’t need. Such a discovery can make it all but impossible to gain approval for further purchase orders.

“Understand what your applications, databases, and end users really need from a capacity and performance perspective before adding new storage,” says Tim Arland, principal consultant for storage solutions at Forsythe Technologies.

On the other side of the coin, if management sees that IT avoided a major storage addition via some diligent homework, it will look far more favourably on the next request for more capacity.

Avoid Vendor Lock-In

Vendors often offer sweet deals for large packages of storage hardware, particularly when their gear is all that remains. The problem here is that you are then tied to that vendor for service contracts and upgrades. That can actually result in longer life cycles and more costs in the end as you are either legally tied in to that vendor or are faced with a substantial up-front investment to replace your hardware entirely.

The best strategy is to own storage gear from a couple of vendors so you can play one against the other. Such a scenario typically results in good deals from all sides.

“Use vendors against each other to create a balance in the bidding process to get your best pricing,” says DeCaires. “Understand pricing models from vendors, be aware of technology complexity that obscures pricing, and carefully examine all the elements of technology packages to ensure you are buying only what you really need and want.”

Avoid The Latest & Greatest

Vendors generally blow the trumpet loudly for their new wares. But these products usually need to have the kinks knocked out of them, and it can be a year or two before they truly are enterprise-ready. That’s why large companies are often very conservative when it comes to storage: They tend to stay a generation behind the development curve in order to deal in only the most stable platforms.

Fibre Channel over Ethernet, or FCoE, and solid-state drives are examples of technologies being touted heavily in the press, but they may not be the best way forward for a budget-constrained SME. However, that doesn’t mean that there are not some newer developments that can add value and help save on storage without entailing too much risk.

“New technologies should be approached with caution,” says DeCaires. “Companies should understand the hype and the reality. New and fairly new technologies that add value and can save on new storage include storage virtualization; thin provisioning, deduplication, and software storage management that delivers utilization monitoring and charge-back features.”

Implement Tiered Storage To Reduce Costs

Storage tier offers a way to boost performance for the most crucial applications while reducing storage costs overall. Tier 1 should be a small amount of the total data set that is mission-critical. That tier gets the best hardware and the highest performance. Another one or two tiers can then be set up using lower-cost disks for the bulk of the organization’s applications.

No comments:

Post a Comment