Best Practices for Implementing a Media Asset Management (MAM) Solution on AWS
Media & Entertainment (M&E) customers seeking to modernize their Media Asset Management (MAM) on Amazon Web Services (AWS) have questions to consider before getting started: Should they build their own solution or invest in a 3rd party system? How can they best leverage AWS services to get the most from a MAM in the cloud? And how can they optimize for both legacy media archives, which may be years or decades old, and new content going forward?
This post dives into those considerations and offers use cases for building vs. buying, best practices for MAM implementations, and a look at the AWS services and partner solutions that support media asset workflows.
What is a MAM and what are its benefits?
The definition of Media Asset Management differs depending on who you ask, and the M&E industry uses the term to describe a wide range of tools and features. Even the name itself is not easily agreed upon – you may hear “MAM”, “PAM” (Production Asset Management), or the more industry-agnostic “DAM” (Digital Asset Management) used to describe the same system. At its foundation, a MAM is defined as a system that manages the storage, organization, and collaboration of multimedia files and their associated metadata. Additionally, MAMs are a key integration point with other components of the media supply chain (e.g., transcode, editing, scheduling, distribution). Ultimately, a MAM helps companies monetize their content by providing insights and secure access to valuable media assets, and optimize their storage costs by managing lifecycle rules for content based on demand.
Should I build or buy a MAM?
The build vs. buy question can only be answered by looking closely at your specific use case. An “off-the-shelf” MAM system is a great option for customers who do not have the time, developer resources, or budget to build a custom solution themselves. The Amazon Partner Network (APN) offers MAM products that can be quickly deployed in either a customer’s AWS account or as a Software-as-a-Service (SaaS) platform. Even in a SaaS model, most customers will “bring their own storage” by leveraging Amazon Simple Storage Service (S3). Customers can use the secure, reliable object storage of Amazon S3 buckets in their own account to house media files, while a SaaS partner manages the backend AWS services that power the system (compute, database, etc.).
Customers who build their own MAM solutions on AWS typically have ample developer resources to design and implement both the foundational backend and frontend user-interfacing elements of the system. Even when building a custom MAM, it is recommended to use partner components wherever commodity solutions exist. Video playback, file transfer / sharing, and workflow orchestration are component examples where feature-rich, industry standard products are available in the market today. Integrating with these existing products can dramatically decrease implementation timelines and allow you to focus on building only the components that add innovation and value to your business. In other words, don’t “reinvent the wheel”.
Vidispine is an AWS MAM partner well suited for M&E customers inclined to build using a combination of custom-developed and partner-provided components in their solution. Vidispine offers core MAM components such as asset ingest and registration as flexible APIs, enabling customers to easily connect them to their internal systems or even build their own frontend application layer on top.
For customer use cases where a light-weight MAM is needed, SaaS solutions available on the AWS Marketplace such as Nomad CMS can be quickly deployed to act as a simple UI on top of existing Amazon S3 storage. Additional features of Nomad include asset metadata enrichment upon ingest via the flexible usage of AWS AI/ML Services, workflow integrations, and an Adobe Premiere panel for seamless nonlinear editing workflows.
How do I best leverage AWS storage services for MAM?
A fundamental task of MAMs is to manage media storage wherever it resides. Many M&E customers use Amazon S3 as the foundation of their media archives for its security, reliability, and tiering capabilities that let them optimize storage costs. For example, infrequently accessed archive assets can automatically cycle to colder storage tiers like Amazon S3 Glacier or Glacier Deep Archive (remaining accessible within minutes or hours) to minimize costs, while low-resolution proxies or other frequently accessed content is stored in S3-Standard for instant access. The newest tier, Glacier Instant Retrieval, gives customers the best of both worlds by providing low-cost storage that is still instantly accessible. Many MAM systems offer the ability to manage content lifecycle policies for Amazon S3 and Amazon S3 Glacier either programmatically or directly via a UI.
To further optimize storage, MAM customers can consider adopting a componentized file spec (e.g., Interoperable Mastering Format [IMF]) in which one piece of content is managed as a group of individual video, audio, subtitle, and metadata files. This provides flexibility for ingest and processing while also eliminating the need to store redundant media, decreasing overall storage footprint and cost.
How do I optimize for both legacy media archives and new content going forward?
A common challenge customers face when migrating their on-premises archives to AWS is that, while the cloud offers opportunities to enhance asset metadata in ways not previously possible (for example, using AI/ML services like AWS Rekognition to automatically tag celebrity faces, and Amazon Transcribe to create searchable transcriptions of content), there are often metadata schemas from legacy databases that contain valuable information customers don’t want to lose in the transition. Similarly, media companies must sometimes consolidate content libraries as part of a merger or acquisition with another company, forcing a reconciliation of disparate databases and schemas.
Amazon DynamoDB is a fully managed, serverless, key-value NoSQL database with a flexible schema that many M&E customers use to store both legacy and new metadata. Existing data can be migrated via the AWS CLI in a zipped and formatted JSON file, via API, or through the AWS Console. New values can be added to your database during the ingestion and analysis process by leveraging Amazon Simple Queue Service (SQS) in combination with AWS Lambda, a serverless, event-driven compute service. MAMs typically abstract these processes from end users by facilitating database value population in the background and making it easily searchable in a UI.
When implementing a new MAM, it is best practice to define a lightweight metadata schema against an ID structure (e.g., Entertainment Identifier Registry [EIDR]) to establish key asset relationships and reduce reliance on file naming conventions that can make it difficult to locate and identify assets.
Where do I get started?
Visit AWS for Media & Entertainment to explore purpose-built solutions, AWS Partners, and see customer stories. Contact Us today to get started on your journey to transforming your media workloads in AWS.