More Scalable and Better Protected than RAID, with Faster Content Retrieval Than Tape

Of the 798 episodes of the BBC’s Doctor Who that have been filmed and aired since 1963, 97 are missing. According to The New Yorker, old episodes were simply taped over when new ones were made: “In the 1960s and 70s, the BBC had customarily wiped tapes after broadcast.” This is unthinkable today. But back then, content creators couldn’t have dreamed of the myriad of ways media is devoured today: Hulu, YouTube, TiVo and collectible DVD boxed sets, to name a few. And yet, how can we guarantee that content we create today is never lost and can be easily located and remonetized? 

Whether it’s streaming previously aired content in-flight, distributing online through Netflix or simply reusing clips as stock footage, careful management of a digital library provides additional revenue and shortens production cycles. To the tireless newsroom video editor, it’s finding source clips of freeway traffic or sunsets over Manhattan to meet tight deadlines for the evening news. For new distribution channels, the sticking point is not brokering new distribution deals, but being able to source the right legacy content quickly and easily.

In many production workflows, content is archived to tape after broadcast or distribution and is not always kept directly available to content creators for future productions. To access that content, producers or editors make a request to an archivist and may wait hours, or even overnight, to obtain requested clips. With modern media asset management (MAM) systems, content can be cataloged and proxies made available for quick review. When the MAM is integrated with a tiered storage and archive solution like Quantum StorNext, high-resolution files can be retrieved with a click of a mouse, even if they’re stored on second- or third-tier storage. 

Tiered storage systems are perfect for managing digital libraries, and disk and tape are economical storage media options for second or third tiers. The long-term issue with content archives built with RAID and digital tape is that they are not self-migrating. That means upgrades to new disk or tape media over time requires manually migrating the files stored on the outdated media. With large libraries, the migration time can be years, and it’s not unheard of for content to be simply abandoned in its original format. Many companies decide that it’s not worth the time and effort for a full migration of assets to a new digital library. 

So if you have legacy content stored on outdated media, it’s best not to delay migration, and any new content you create should ideally be stored in the most technologically sustainable manner possible. Think as if you were a brand-new media company building an infrastructure for your first piece of content now. How can you be sure that the storage and distribution solution you start out with today will hold you in good stead for the foreseeable future? Enter object storage, enabling a forever disk achieve that can be part of a 100-year workflow.

Even though it’s been around for more than 20 years, object storage is only now entering media workflows as content owners grapple with petabyte or greater content archives. Object storage is best known as the technology that powers the world’s largest public cloud storage. Its increasing use has recently made it affordable and practical for companies with outsized storage needs. Because it affordably combines disk-speed access with long-term durability, more media companies are turning to object storage for long-term storage of digital assets. 

Object storage’s workflow breaks from the traditional hierarchical file-system architecture, which inhibits flexibility in global access. Instead, object storage assumes a model that can be compared to valet parking. Think of objects as your car, and the ticket you’re given at the valet stand as the object’s ID. You may not know where your car is parked while you’re at dinner or the theater; all you know is that it is driven to the curb once you produce the ticket you were given. Similarly with object storage, there is no limit to the physical distance of the data you need to retrieve, and it doesn’t matter. All that's relevant is that a clip of a particular Jeopardy! question is available to you right when you need it, and can be pulled back and reused immediately.

The object storage technique for storing data also helps to prevent disruptions in data integrity that can plague traditional RAID systems. Called erasure coding—not as ominous as it sounds—this process involves converting data into an algorithm that is spread throughout the object storage system. As a result, it shortens time to recovery after common disk-related failures that can disrupt data integrity or access performance in traditional RAID systems. Moreover, this diffuse model allows for equal, swift access to content on a global scale.

As a disk-based storage, object storage can start streaming content immediately versus tape, where there’s a lag while the tape is retrieved. But unlike traditional RAID disk, it’s more scalable and offers better protection at a lower price. Plus it makes migrating to new disk technology as easy as replacing drives on RAID failure. It’s a perfect fit for digital libraries that are massive and need to be long-lasting. 

As with all technology, there’s no ironclad guarantee of longevity with object storage. But its automatic migration alone offers invaluable efficiencies for content owners that make it difficult to pass up. So do the next generation and the one after a favor—make your move to object storage now and you’ll be ready to remonetize your content for decades.

Janet LaFleur is senior product marketing manager at Quantum