FEED Winter 2022 Newsletter

might be a good time to start looking to some of those earlier-mentioned efficient solutions now appearing on the market. SERVERLESS SUSTAINABILITY “Let’s say I want to move a legacy MAM system that requires 13 separate servers and a heavy database, with each of those servers needing to operate all the time – whether I had it on-premises or in AWS,” Carson explains. “Now compare that to an application like Ateliere, which was designed to be serverless from the very moment it was born.” Ateliere offers an entire service architecture, without heavy virtual machines and operating systems working away behind the scenes to support the application. “The end service doesn’t care which machine it’s running on, as long as there is actually a machine available for the computation to run. Ateliere can start that service, execute the tasks as necessary (whether that’s metadata ingestion, a transformation of media or even a distribution job), then turn that service off as soon as the job is done. “So we’re not consuming additional power or computation, or cooling at the time services are needed. To me, that’s a seriously solid solution, inarguably contributing to a greener media industry.” That’s the power consumption question answered. But what about that growing issue of duplication? “Ateliere offers a unique feature called deep analysis,” says Carson. “This includes a deduplication of content function as it travels through the cloud, meaning it takes those video archives and reduces the

amount of storage needed, which cuts the number of spinning disks and eventually slashes those archives by 70% or more by the time they’ve entered the cloud. Ultimately, this creates a simpler path for making distribution versions – so we’re consuming less computation power to then make versions out of that deduplicated library.” Ateliere also has a function that converts consolidated libraries to IMF. The Interoperable Master Format is a SMPTE standard that has been designed to store content in multiple formats with maximum efficiency. It works by creating a single master version of a piece of video content, whether it’s a movie, TV episode or commercial. Once that master has been made, versioning can be completed by simply creating files that log all the differences. “It’s our sound belief that IMF is the most efficient and effective way to store and utilise your archive going forward,” Carson concludes. WHAT NEXT? There’s undoubtedly a lot of work to be done on M&E’s part in order to see through its hazy, self-made veil of pollution. However, there are some easy steps for greening your workflows if you wish to take action sooner rather than later. Contact an Ateliere workflow specialist today to learn about how you can reduce your cloud storage footprint and take a step forward for sustainability: ateliere.com/ ateliere-connect/

with facilities built from the ground up, it automatically has less power input. And it’s designed to run at a higher temperature, while keeping the necessary cooling bandwidth. “They also aren’t limited to using only what’s available at one location. Take Microsoft for example, who put a data centre in a storage container and dropped it in the ocean to keep cool.” Data centres are normally swollen with idle, energy-munching files. This is a particular issue for those in M&E, with a massive file footprint due to what it takes to store, manage and distribute titanic chunks of media. Accelerating this file bloat is the issue of duplication, which comes from content owners striving to meet the exponential demands of consumers wanting to watch all the latest shows. These demands result in the creation of versions to meet numerous compliance requirements, languages and localisation needs – with one title potentially generating hundreds of additional versions. These scenarios could push a storage environment to its limits – and moving all that material into the cloud can start to be an expensive proposition, in addition to its sustainability implications. This ATELIERE’S DEEP ANALYSIS CAN DEDUPLICATE DATA, SLASHING ARCHIVES BY 70% OR MORE

@feedzinesocial

Powered by