The emergence of 'content depots', large repositories of digital content amassed and organized for distribution, is changing how enterprises are consuming disk storage space given the high volume these depots require.
While the traditional trend of migrating from a direct-attached and internal storage system to SAN and NAS has mostly continued, Richard Villars, vice-president of storage system research with Framingham, Mass.-based IDC, said about two years ago "we were finding a large pool of storage being consumed that was running counter to [that]."
"All of a sudden there was this huge spike in demand for what was traditionally thought of as older systems architectures," said Villars.
Looking a little closer, Villars said spending in that area was being made by organizations like Amazon, Facebook, MySpace, and AOL. And, as a result, a new market for companies building content depots had emerged that didn't exist five years ago.
Network-based, starting at a petabyte of capacity and reaching "tens, twenties even hundreds of petabytes", content depots are primarily used to store file-based data like images, movies and records, rather than transaction-based data produced by credit card systems.
But content depots are not solely catered to consumer needs. They are deployed in areas like oil exploration for storing large pools of seismic data, and life sciences for genome data or protein sequencing. Basically, said Villars, any corporation whose business goal is to collect and store huge amounts of data for an extended period of time would use a content depot.
And the pool of organizations using content depots is expanding as certain industries adopt an online delivery approach. Media and entertainment, for instance, is increasingly offering digital content delivery, and telecommunications providers are offering video delivery and photo storage, said Villars. "We are seeing traditional companies setting up content depots as well and transitioning into that same type of model," he said.
But besides the obvious adoption driver of explosive data growth and increasing digitization of information, customers are figuring out ways to build content depots cheaply, said Villars.
Disk drive manufacturers, in turn, are responding to the re-emergence of direct-attached storage. "What you're seeing really in 2008," said Villars, "is all the major players, EMC, Dell, HP, IBM were suddenly realizing there was this huge consumption model" and are attempting to pursue this space.
"The IT players running these businesses are really the ones setting the tones for things like power consumption and virtualization and density, because these are people who are pushing the edge of the envelope," said Villars.
The IT administrator tasked with building a content depot, said Villars, needs to recognize that this storage environment is architecturally different. And, they should have a good knowledge of cluster file systems and really consider the underlying operations because, he said, "you have to think about what it means to run 40 petabytes of capacity inside the data center. You have huge concerns around power consumption, heating and cooling, optimizing system design."