Post Magazine

November 2013

Issue link:

Contents of this Issue


Page 39 of 51

post positions Managing video content growth on a budget By Daniel Jaramillo, Vertical Market Specialist & Hossein ZiasHaker Senior VP/Business Development & Alliances Spectra Logic Boulder, CO M edia and entertainment organizations of all kinds are struggling with a common problem: explosive video content growth and shrinking or static budgets to manage it. The progressions in filming quality for movies, such as Avatar and The Hobbit at greater resolution and higher frame rates only proves that this problem isn't going away, but is going to increase tremendously. Production at even frame rates such as 120 and 300 frames per second is on the horizon, which will result in an even larger storage requirement — much larger — as much as 30x larger.This doesn't include video and audio effects done in post; migrating video assets from legacy technology such as Betacam to digital files; the multiple copies of rendered files; and freelance work that also consumes large amounts of space on storage systems. And it's not just limited budgets that bring a challenge; it's also the resources required to manage it. The typical reaction to this problem? Buy more spinning disk external hard drives. This reaction doesn't solve the problem. It simply postpones dealing with the challenge, or even makes it bigger. So, is it possible to solve this problem costeffectively while still efficiently managing the assets? Yes. The mix of technologies that make it possible will depend upon your workflow, your disaster recovery (DR) and your business continuity (BC) strategy. In most environments, having a balance of relevant storage technologies (flash, disk, digital tape and cloud) is the best way to approach solving this problem. In order to meet the challenge, it is critical to understand each technology, its future dynamic, and its relevance to the value of the assets. 38 Post • November 2013 Let's start with flash-based storage subsystems (flash as a system, rather than the basic USB thumb or external TB drives you can purchase at the electronics store).Transactional and computational markets traditionally held by the high end of the spinning disk-based solutions are being taken over by flash-based storage for greater performance. In the media and entertainment world, this means editing systems, gateways for screaming fast streaming, or any case where high input/output operations per second (IOPS) is required. Flashbased technology has gone from 35 microns to 22 microns and now, thanks to Intel, to 14 microns. This means higher densities and capacities will be coming to the market soon. The above dynamic and the shrinking die sizes continue to drive the cost of flash technology and put more and more pressure on the spinning-disk market. Spinning disk is known and used for its random access attribute, high performance, and ease of use. This market enjoyed both high margin and high volume aspects, which up to this point, lead to its success and market growth. But spinning disk is now facing technological hurdles affecting its density/capacity growth, reliability and cost advantages. As a result, we see the market shifting. Spinning disk is being impacted from two sides: flash on the high performance/high margin side, and data tape on the volume side. The truth is, spinningdisk technology wasn't designed for the capacity and long-term retention required by most environments today. That said, spinning disk still has relevance and effeceffec tiveness in production environments: as high IOPS solutions for editing and VOD. Streaming-optimized solutions such as Spectra Logic's nTier Verde meet the short- to mid-term retention needs of ingest, digital parking lot for semi-static content, or simply as utility storage and archive cache. These days, media and entertainentertain ment organizations are keeping their video content meant for long-term storage on data tape. Cost is a big factor in this, but with the evolution of the technology, data tape has also become more reliable, dense, scalable and energy efficient than any other storage technology available. A recent study showed that data tape has become over 700 percent more reliable than tape technologies of a decade ago. Data tape also has a roadmap for reliability and scalability, and the great thing about LTO is that organizations can cross gen- erations with the ability to read back two generations and write back one generation. This allows organizations to scale and save money without having to do forklift upgrades with every new generation's release. With the speed and capacity of data tape, media and entertainment organizations can manage the cost of their storage in a multitier storage model where spinning disk is deployed only for active and semi-static content, but uses LTO for active archiving of their static content, ultimately saving thousands to millions of dollars. Factor in the emergence of LTFS and the use of proactive monitoring tools such as Spectra Logic's Media and Drive Lifecycle Management, and digital tape has become the most cost-effective and safest place to store video content. Over the past couple of years, the concept of the cloud has become a popular topic. However, the cloud isn't a new concept. It's been around for years — we used to call it "managed storage." As higher speeds of Internet bandwidth have become more affordable, the idea of cloud storage has become more feasible, even for budgetconscious and resource-deprived organizations. Common uses of cloud storage are general purpose storage, distribution, and collaborative work where multiple editors or groups work on a common project. Once the collaborative work is finished, the project is typically moved off the cloud. However, when it comes to managing one's assets, there is a strong resistance to deploying a cloud-based solution unless it is a private cloud. Progressive companies, like Spectra Logic, are now offering tape as a cost-effective private cloud, serving as a deep storage repository to manage the ever-growing content that M&E companies are facing. There is a school of thought that a public cloud may offer the final insurance in case of a major disaster. Those who do consider public must carefully consider the following: cost and price stability, the process and cost of bulk recall, vendor lock and switching costs, recall performance vs. cost of WAN, availability (downtimes), security (of content), the viability of the provider, legality, and control. Solving the problem of storing and managing video assets in a world where content is exploding and budgets are shrinking doesn't have to be complicated.The technologies mentioned here are effective when used purposefully and with the proper understanding of their roles in an M&E environment.

Articles in this issue

Links on this page

Archives of this issue

view archives of Post Magazine - November 2013