TV Technology pp14-16 Sept issue
https://issuu.com/futurepublishing/docs/tvt465.digital_september_2021?fr=sM2QyYTM4NzgxNjI#
As the cloud is increasingly applied to media production, the lines between ground-based and cloud-based media environments are becoming blurred. The asset management system provides the glue between them but perspectives on the best way to use it are changing.
“Even if you think you know what cloud storage might cost be sure you’re updated on all the recent facts, variables and combinations of services before making a move in either direction,” advises Karl Paulsen, CTO of Diversified. “Like SAN storage, cloud is an evolving and frequently changing environment.”
We asked experts at a number of vendors to offer guidance toward the migration of assets, the cost of storage and MAM’s crucial role in orchestrating this. All agree that no matter the long-term advantages of cloud production a ‘big bang’ move is unlikely to pay dividends.
Big bang or piecemeal approach
“Anyone involved in technology migration projects will probably tell you that it’s much easier building systems from scratch than migrating existing workflows to new systems piecemeal,” says Raul Alba, Avid’s director of solutions marketing. “But the reality is different. From our deployment experience, we’ve found a ‘tabula rasa’ approach isn’t always right because a return needs to be made on past investments, or because entirely-new workflows will cause too much operational disruption. This means hybrid workflows will be important for several years, allowing every company to transition to the cloud at their own pace.”
A hybrid scenario implies that the media lives, and is processed, both on-prem and in the cloud. In some situations, hybrid operations will be required regardless as it may be more cost effective to work on-prem before assets are moved to the cloud.
“Migration could begin with archive or short-form digital workflows, then move onto complex workloads once the organization understands the new ‘physics’ of working in the cloud,” says Lincoln Spiteri, VP Engineering, Dalet Flex. “Users should be aware of both the advantages and limitations that come with working on the cloud. Lessons learnt should be used to optimise workflow.”
SDVI’s Simon Eldridge similarly recommends taking a specific use case and getting that up and running before starting a process of iterating and optimizing.
“Don’t try to do everything at once – iteration leads to transformation,” he says. “The value of that quick win, coupled with the experience the team will gain from that initial migration will accelerate the process of implementation for the next use case, which will be logically selected based on its returned value and business impact.”
Tedial say cloud has had a major impact on traditional MAM workflows and makes the case for a Media Integration Platform. CTO Julian Fernandez-Campon explains that this allows broadcasters to have the same workflows running on-prem or in the cloud, as the applications are integrated and multiple storage locations will be transparent to the platform.
“A hybrid cloud deployment using a Media Integration Platform provides a low-risk transition by moving operations in-line with business needs allowing them to pivot between local and remote operation depending on requirements.”
Cost considerations
The cost of cloud can be a minefield. Among other things, costs vary based on where the physical data centers are geographically located. “It’s a bit like going to the smorgasbord,” says Paulsen. “Many items are à la carte.”
Fees include monthly access, retention time, storage volume and/or the use of inherent capabilities of the store itself. There are hidden costs for deletion of content.
Paulsen says, “Use it once and quickly—a short term ‘put it there and take it back out’—and you’ll have one price. Leave it there for a lengthy period of time—another price. Need rapid access to an archive? You can watch previously expected low-budget costs to take off like SpaceX launching multiple satellites one at a time.”
It stands to reason that M&E needs to have a good understanding of the assets they ingest and how these will be used. This information drives decisions about how to optimize storage in terms of hot, cold, and archive storage tiers, which directly impacts costs.
“MAM platforms should fully support cloud storage to balance out storage costs and timely media access,” says Spiteri. “For example, low-bitrate proxies enable low-cost access to archives. Source assets should be pushed to archive storage as soon as possible while lower bitrate mezzanine formats are a good option for editing.”
Geoff Tognetti, SVP & GM Content Management Business Unit at Telestream offers four questions each organization has to address to “accurately estimate” cloud storage costs:
- What percentage of your existing content footprint should be migrated (typical cloud migrations we’ve seen range from 30-75% of current LTO storage, not counting copies).
- Understand your cloud provider’s tiering policies and how those map related to your lifecycle rules on premise.
- What media format do you want to use for the assets that you store in the cloud, i.e. your production format, a separate mezzanine format, proxies, etc.
- Estimate the impact on your current production workflows (number of drives/data movers dedicated to migration, off-hour scheduling, etc.)
The essence of metadata
Metadata plays a key role in giving content owners visibility of their assets to be able to quickly locate relevant media before it is moved around.
“Having media stored anywhere is pointless if you can’t find it quickly when you need it,” says Alba. “If you can easily find and retrieve only what you need, you’re saving both time and money.”
Having an adjustable data model, “with time-based metadata and a sophisticated search interface is key to making media management systems efficient,” he says.
In addition to object or technical metadata that is mostly considered in a cloud storage environment, you also have to consider other descriptive metadata such as transcripts, and visual logs.
“Cloud access makes it easier to generate and preserve this data, and the ability to relate it with media files can save a lot of time in post,” informs Russell Vijayan, Business Manager, Digital Nirvana. “While such metadata is enhancing process speed, reducing time to market, and improving overall efficiency, there’s still no considerable data available to determine whether the current price points of storage could be justified. For the most part, high-volume users with a clear cognitive metadata plan and usage would stand to benefit.”
“Strong” metadata gives organizations the ability to “track and understand the value of assets, drive asset lifecycle to help manage cost and ensure that correct assets are located and moved through production workflows,” according to Spiteri.
Generally speaking, the more metadata that can be added automatically the better. The most important point, says Fernandez-Campon is the concept of Metadata Aggregation from multiple sources. “This provides a standardized metadata model to combine multiple metadata sources and provide business value.”
Eldridge asserts that as soon as you move to the Cloud, the data universe of “what can be automatically generated and collected by the system just explodes.”
Contextual, time-based metadata is generated by QC or ML tools that have analyzed the content, he says. That metadata can assist operators looking to segment or make compliance decisions. Time, cost, and performance data can inform decisions around which tools you use and for what purpose.
“Tagging data can be used to assign costs at a granular level based on process, asset, project, or network. Rather than being considered as a sidecar to the content, metadata becomes a critical driver of your supply chains and an output that helps you further optimize.”
The evolution of MAM
Eldridge is specifically referring to a “cloud-native supply chain model” which is the natural evolution of MAM, according to SDVI. It argues that traditional MAMs were built around fixed capacity infrastructure and heavily customized around use cases – therefore unfit for the agility required of cloud production.
“Today’s media enterprises don’t want to worry about capacity planning or constraints, and they want to pay only for the services they consume,” he says. “Perhaps most importantly, they need to be able to predict cost easily in order to support good decision making. They need to allocate those costs accurately to fully understand the profitability of the products they make.”
Dalet’s Spiteri, agrees; “MAM systems are morphing to media supply chain management systems, where a wide range of media types are managed with a wide range of automated workflows. Once these systems sit in the cloud, they become collaboration platforms, expanding their utility.”
He adds, “Modern MAM systems cannot always fulfil every use-case and this is where it’s important that they offer integration tools and open APIs to create a truly business-driven ecosystem.”
Avid’s Alba acknowledges that the term ‘MAM’ can be associated with “complex, expensive and hard to maintain systems” but says that’s not the case in practice. “In fact, our users tell us these solutions are more important than ever both because of the amount of media and metadata being managed, and the number of outlets to which it needs to be delivered. MAM is an essential part of making media production workflows efficient.”
A further point worthy of consideration is that cloud relegates the once painful, time-consuming, and labour-intensive process of media migration to history. That’s a misconception asserts Tognetti. “Just as moving from one LTO version to another was inevitable, so is moving from one cloud vendor to another. Cloud storage providers have no incentive to facilitate any migration outside of their products.”
The solution for this, he says, is to look for agnostic content management provides tools “to automate the migration process for all vendors” through cost-effective, and non-disruptive workflows while serving the lifespan of content.
M&E organizations need a scalable and elastic platform that allows them to rapidly build, deploy and iterate on its supply chains as business needs evolve. Call it a MAM if you like but asset management has never been more important.
No comments:
Post a Comment