• by Matt Berry , February 15, 2010

Last year, high-profile online video platform (OVP) announcements were everywhere. Think Netflix's streaming partnerships, Best Buy's partnership with CinemaNow, and Disney's Keychest.  All of the initiatives were designed to securely bring content to an assortment of different connected devices (my previous post on connected devices is here).

With these big video announcements, we keep hearing the unavoidable buzz term "cloud computing" again and again. 

Obviously, video storage and delivery can be complicated and expensive.  So it's not surprising that media companies are considering cloud as a way to expand their businesses and save money.  Let's briefly explore cloud, and whether or not it's a prudent solution for media companies. 

Cloud computing, also commonly called Software as a Service (SaaS), can refer to different things, but basically it means applications that reside on the Internet instead of on your computer. 

The first, and perhaps most obvious, thing to evaluate when considering cloud is this: Internet access is required to get any data or media stored inside the cloud.  If you don't have access to the Internet or your connection goes down, you can't reach the content or data.  There are also inherent security concerns, though media is often as exposed on local servers as it is online. 

At the moment, the cloud decision really turns on scalability, more specifically on the current business model that governs companies in the SaaS remote storage space.   

Cloud companies traditionally charge on a byte in/byte out fee -- so for large, tier-one media companies like movie studios, a storage cloud service does not scale.  

A concrete example of this: If Studio A has a copy of "Harry Potter" that it wants syndicated onto the Internet, then it will have to transmit a huge file (50 gigabytes or more) to the transcode cloud.  The finished product is a file that is 200 megabytes.  Because Studio A is being charged a byte IN charge, this would not scale due to the size/cost ratio of the incoming file.  

On the flip side, Studio A could pay the cloud vendor that is doing the transcoding to avoid the byte IN fee, and then be exposed only to the byte OUT fee.  That arrangement scales much better, but storage fees have hovered around 50 cents per gigabyte per month.

So, let's do the math and see if this makes sense for Studio A.  The company has a digitized master library with 50Mbit content roughly 50 terabytes in size -- which comes to somewhere around $20,000-30,000 per month for storage costs alone.  Those fees would eliminate the savings generated by dropping the byte IN model.   

In other words, companies in the storage/transcoding business will need to rethink their models before they can sell services to tier-one media companies.  

That said, the ability to scale Web services across many processors in the cloud has huge potential in the short term, without changing the models.  This is why you are seeing great advancements in virtualization and newer intelligent load-balancing systems directed at dynamically driven services.   

In general there is a huge future for this technology, and that's why nationally based hosting services are tooling up to offer cloud-based "computing."  Having hundreds, if not thousands, of processing cores on standby that anyone can access on an as-needed basis is like having a supercomputer accessible from your home.

Expect cloud-based computing to continue to grow and become more prominent as more devices allow users to connect from anywhere and everywhere.  It's not quite there yet, but as with media delivery itself, these business models will mature to serve market needs.

Depending on your application of the technology, cloud computing can have significant benefits.  I expect cloud to continue to grow, and become more prominent, as the business model matures.