NetApp StorageGRID for media workflows

With the StorageGRID single global namespace, it doesn't matter where or how an object is stored and protected, it's available to any user at any site at any time. With massive scalability across this namespace, StorageGRID frees your media applications from the task of finding and moving data among volumes, systems, and sites. StorageGRID enables you to establish highly granular, flexible data management policies that determine how data is stored and protected throughout its lifecycle. When developing and enforcing policies, the object store examines a wide range of characteristics and needs, including performance, durability, availability, geographic location, longevity, and cost.

StorageGRID is not a media asset management system. It's a data management system that offers sophisticated policy-driven functionality to support your MAM systems. Policies can be enforced at ingest, at rest, on read, after object attribute update, at object disposition request, or when information lifecycle management (ILM) policies change. StorageGRID policies can initiate object replication, erasure coding, tiering, and archiving to the cloud. This storage management, monitoring, and migration functionality delivers an improved experience for all the workflows in your media enterprise. MAM systems support specific production or delivery workflows in an enterprise. An object store confirms that the data that those systems need is in the right place at the right time, and that it's easy to access that data through the global namespace. By erasure coding, moving, or copying objects to sites to meet production and distribution access demands, StorageGRID relieves your MAM systems, users, and administrators from manually performing these tasks. When an object is stored, it is seen and accessed as one object by all clients, regardless of where it is or how many copies exist. With object store support, MAM systems can pass key metadata along to other systems and workgroups by storing it with the object along with the media file data, or "payload."

Erasure coding is a method of data protection in which data is broken into fragments and encoded by mathematically combining the fragments into a greater number of data pieces composed of multiple copies of the original fragments. It is then dispersed across different locations or storage media. Erasure coding offers better overall storage efficiency than replication, with equivalent durability and availability. In an object store with geo-dispersed erasure coding, data can be retrieved from a subset of the data pieces. With ILM policies, every object can use a different erasure coding scheme. StorageGRID offers a variety of erasure coding schemes that you can choose or change to at any time. StorageGRID appliances leverage an additional layer of erasure coding. The appliance uses Dynamic Disk Pools (DDP) technology to provide drive failure resilience at the node level. This 2-level erasure coding allows object access even during temporary or permanent failure of one or more drives, nodes, or sites. It also reduces network consumption during rebuilds, giving StorageGRID up to 9 9s of availability and 15 9s of durability.

Achieving ubiquitous content access often requires storing copies of the same asset in multiple places. This approach increases complexity, and it also requires additional storage capacity. With StorageGRID, you can minimize the capacity that is needed while optimizing for data resilience, production access, and distribution requirements through policydriven automation. The policies that you set to move and copy objects can also include deletion criteria. For example, copies of objects can be automatically deleted after a set period of inactivity or after their distribution rights have expired, reducing capacity requirements. The value of content changes over time, as does the cost of storing it. High-value content requires highperformance storage to keep it readily available. As content ages and is accessed less often, policies can automatically reduce the number of copies, move to an erasure-coded protection scheme, or tap into hybrid cloud by moving that content to less expensive public cloud storage. Management and maintenance of file copies, migrations, and deletions are driven by the StorageGRID ILM policy engine.

StorageGRID is built to fit into a multiworkload, multitenant architecture. Whether it's using S3 as a backup target, active archiving from your MAM, or capacity tiering via FabricPool or other tiering solutions, StorageGRID enables you to painlessly run multiple applications. In fact, StorageGRID offers robust multitenant features for security, access control, and traffic limiting, allowing everything from financial, DevOps, and media archives to coexist in the same namespace while maintaining secure, authorized access. As your use cases scale, capacity can be easily allocated and put into service immediately. Maintenance and expansion procedures can be done without interruption or downtime, across generations of StorageGRID nodes. The latest StorageGRID nodes are built with object as primary storage in mind. This means high-performing ingest and egress of your media files in seconds, not minutes. The content in your repository will live longer than the media that it is stored on, and StorageGRID can greatly reduce migration challenges and costs in the years to come. The policy engine will drive the process of moving millions of objects from aging media to newer, lower-cost, higher-capacity media in the future. Upgrades, expansions, decommissioning, and refreshing activities are all nondisruptive in the StorageGRID environment.