In its latest Digital Universe study, International Data Corp. (IDC), Framingham, MA, estimates that the worldwide volume of digital data grew by 62 percent between 2008 and 2009 to nearly 800,000 petabytes (PB). This year, IDC projects what they’re calling the Digital Universe will grow to 1.2 zettabytes (ZB) in 2010 and reach 35 ZB by 2020. A zettabyte equals 1 million PB.

As reported by Tech Target, that data will be stored in 25 quintillion (that’s 25,000,000,000,000,000,000) storage containers (packets, files, images, records, and signals) of various sorts. Today, your data, which probably amounts to hundreds of gigabytes (GB) or even hundreds of terabytes of information, is stored mainly in files and records. If you are a really large organization, you may have one or a few petabytes of information.

There is a cost to storing this information. The industry prides itself on driving down the cost of data storage to pennies per gigabyte (GB) or less. Multiply those pennies per GB by the amount of gigabytes that go into 1 TB (1000x) or 1 PB (1000×1000) of storage, and it starts to add up to real money, enough to impact your balance sheet for sure. Fortunately, there are ways to reduce the cost of storing this data.

So, what can you do to reduce or avoid this expense? Short of throwing data away — which isn’t a bad idea, as suggested here before — there are three storage technologies that can help to slow the growth of your storage and put off the need to buy more: (1) thin provisioning, (2) deduplication, and (3) automated storage tiering.

Thin provisioning is a form of overbooking storage for applications. Applications generally want to reserve more storage than they actually use. Thin provisioning is a way to give the application the storage it really needs now. Should it need more later, provide it then. In most cases, you can put off buying more storage for quite some time. And since the cost of storage trends lower, whatever you buy later will cost less.

Organizations invariably store multiple copies of the exact same data, especially when it comes to data protection. Deduplication technology identifies duplicate data and, instead of storing it again, it leaves a small pointer to where the original copy of the data is stored. With deduplication, organizations that once could store a few days of backed up data can now store a few weeks using the same amount of storage capacity. That comes to a lot of storage capacity you’re not buying.

Finally, automated tiering monitors the organization’s use of storage and automatically moves the data from costly, high-performance storage to lower-cost storage when the data is no longer being frequently accessed. This happens automatically, and, of course, the system can always get it back fast.

Two companies, EMC, which provides FAST (fully automated storage tiering), and Compellent, which provides highly virtualized storage and automated tiering, made big splashes this month, each with a series of announcements aimed at helping companies to deal with the explosion of data to be stored. IBM, HP and others also provide technologies to slow storage consumption.

The bottom line: You’re buying more storage than you need. ###