It’s pretty easy to take new-gen video for granted. We haven’t always had IMAX theaters, 3D movies and augmented reality applications; this has only happened in recent years behind powerful servers, virtually unlimited storage, super-fast networks, lean code and ingenious new applications.
Entertainment experiences that audiences enjoy today represent massive technology advances in HD, 3D, 2K, 4K, HFR, UHDTV, and others. Better video resolution, frame rates and color depth also demand more from IT infrastructures. Technologies that are transforming data centers in other sectors are enabling television and video teams to produce, edit, finish and deliver clearer, crisper and more lifelike content faster and at lower cost.
In the early days, it took $40,000 workstations from companies such as Sun Microsystems and SGI to produce movies such as “Toy Story” (1995) and others like it—and even those weren’t particularly fast in rendering all those video files. Not anymore; a long list of new-gen laptops now are powerful enough to produce high-quality video for movies, video games and AR applications.
In this eWEEK Data Point article, using industry information from Excelero VP of Corporate Marketing Tom Leyden and Pixit Media VP of Global Marketing Bob Murphy, we offer eight data center technologies that are enabling film and video creation teams here in 2018 to innovate, all the way from production to the streaming of films to your living room.
Data Point No. 1: Ethernet Storage Fabrics: 25GbE, 50GbE or 100GbE
The advent of 4K and 8K video, and the demand for 4K/8K uncompressed playback by collaborators who are creating feature films, has forced organizations to reexamine whether their Fibre Channel networks can keep up. Today, as ethernet storage fabrics (ESFs)surpass Fibre Channel speeds geometrically and at much lower cost, ESFs are increasingly prevalent at media & entertainment firms. Even organizations that upgrade to 25GbE before considering 50GbE or 100GbE find they can leverage the speed, flexibility and cost efficiencies of ethernet versus other transport approaches.
Data Point No. 2: Cloud-Enabled Infrastructure
With constant tight project deadlines and competition for new projects, studios often need to spin up IT infrastructure quickly to get back on schedule or add capacity for new project wins. Augmenting their own infrastructure by “bursting” to the cloud, gaining access to more compute power at peak times and accelerating large or behind-schedule projects, is an increasingly common component of film and video production. Storage in remote clouds is playing an important role in enabling collaborative workflows and archiving (Coughlin Associates), with 48 percent of professionals in the firm’s 2018 annual survey using cloud-based storage for editing and post production.
Data Point No. 3: High-Performance Parallel File Systems
Like many industries where IT teams must handle either very large files, or very large sets of small files, media and entertainment workflows are built around scale-out parallel file systems that can provide fast access. Newer solutions that incorporate high-performance NVMe SSD’s along with parallel file systems are one way M&E players are enabling scalability along with suitable high performance/low latency storage.
Data Point No. 4: Transparent Data Tiering
Video projects shot in the past few years can easily require more storage than decades of older assets combined, yet assets rarely can be deleted. Transparent tiering solutions move assets from expensive production storage to low-cost archival storage and provide analytics and insight into the asset and help to move data to the tier that meets the cost and accessibility requirements. Tiering solutions provide a single view of all assets with integrated search, often assisted by AI and machine learning, allow teams to find assets quickly for further monetization.
Data Point No. 5: Advanced Analytics
Analytics tools now can provide insights on who, what and how data is being used, so IT teams can decide what data, film and video content or ancillary files can be tiered and what data can be removed--something that was largely unknown until recently. With this knowledge system, administrators are better prepared to control the growth of data and resource consumption before it becomes critical and make intelligent, informed decisions moving forward, instead of simply always buying more storage.
Data Point No. 6: Multiprotocol Support
Given extensive legacy applications, media and entertainment organizations often meet storage requirements by building multi-tier storage infrastructures and implementing hybrid cloud strategies. To ensure that those applications integrate with new storage solutions, multi-protocol support has become a key requirement. They migrate data from POSIX (NFS, SMB) protocol-based systems to object and cloud protocol-based systems (Amazon S3), and once migrated, assets are stored in industry standard formats; applications can read data directly from object and cloud storage, without having to re-migrate assets back to primary storage.
Data Point No. 7: AI in the Studio
Studios are generating exponentially increasing amounts of digital assets that can be reused and monetized decades into the future. Locating specific assets for reuse through manual techniques is becoming increasingly problematic, risking that valuable assets are lost forever. Enter AI and machine-learning technology, which can automate asset descriptive metadata extraction, allowing users to perform fast content searches for automatic analysis and monetization of millions of visual assets.
Data Point No. 8: Software-Defined Infrastructures
New film and video production formats are driving processing and throughput requirements by orders of magnitude every couple of years, and media organizations have no choice but to keep pace so they can monetize content assets as long as possible. To date, this meant studios had to periodically rip and replace expensive proprietary hardware systems. With a software-defined infrastructure, studios can keep up with ever-expanding requirements in server, network and NVMe (non-volatile memory) storage technology without having to disruptively tear out gear and start all over again.
If you have a suggestion for an eWEEK Data Point article, email [email protected].