Calculating Costs: Spending on Smarter Storage

PDF DownloadDAS vs. NAS vs. SAN. It’s not a German lawsuit. It’s an evaluation that technology managers must make when deciding which architecture should manage company data.

In direct-access storage, or DAS, drives are connected directly to the computers whose data must be preserved. If you haven’t moved to a storage-area network (SAN) or to network-attached storage (NAS), you probably already know what a headache DAS is to manage. But you may not realize the cost impact of having even a small portion of servers still using DAS.

A SAN, the most sophisticated option, is also the most expensive to set up. But total lifetime costs for a SAN are far less than for direct-access storage. That’s because DAS is far less reliable. It conks out more often—about 70 hours per year, on average, compared with a SAN’s average annual downtime of just 24 minutes. And time, after all, is money.

NAS costs typically fall between the other two architectures, both in initial and ongoing expense. But again, figure in likely availability before deciding. The difference in average uptime of NAS versus SAN may seem small, at 99.9% compared to 99.995%. Yet in a 24-hour by 7-day operation, three nines equates to almost nine hours of downtime per year.

Use NAS in back-office applications, such as financials, where some downtime won’t kill you, advises Tom Pisello, chief executive of Alinean, an Orlando, Fla., consulting firm that specializes in figuring the value of technology.

SANs, on the other hand, are best for critical applications, where downtime will immediately cost the company income. MGM Mirage, the $4 billion casino company, for example, runs a SAN to manage the 7TB of data it keeps on gamblers who play at its 16 properties.

“What’s on a SAN at most companies is family jewels,” Pisello says.