In the face of larger file sizes and ever-higher resolutions such as HDR, 4K, 8K, 360-video and virtual reality, storing petabytes of media is a significant challenge that’s all-too-often a cost vs performance trade-off. What’s needed is a scalable, intelligent and cost-effective storage foundation – and one that doesn’t just hold a media archive but provides anytime, anywhere access and a way to extract more value from an asset.
Legacy systems fall short
As capacity requirements grow exponentially, traditional storage area network (SAN) and network attached storage (NAS) disk-based solutions—which are often based on expensive proprietary hardware—not only become cost prohibitive but also insufficiently scalable. By nature of their design, these legacy systems create silos of data as storage capacity grows, making it difficult to provide a reliable platform for geographically dispersed teams working collaboratively and ultimately hindering seamless access to all content. Multiple islands of storage also result in greater management complexity, cost and risk, and scaling requires continuous upgrades to larger infrastructures.
Similarly, because of latency and management issues, tape storage systems that have been relied on for so long simply can’t keep pace with the demands of today’s media workloads.
What’s needed is a way to seamlessly and cost effectively scale to hundreds of petabytes without interruption or downtime and ensure that content is readily accessible anywhere and anytime.
Object storage answers the call
At its core, object storage is an approach to addressing and manipulating data as discrete units called objects. Rather than organising stored content into a hierarchy with directories and sub-directories, object storage uses a flat address system in which data is stored as objects along with metadata tags and a unique identifier.
These objects are stored in a flat address space, eliminating the hierarchical structure used in file storage and making it far easier to locate and retrieve data across regions. This flat address space also helps with scalability. By simply adding more nodes, users can scale to petabytes and beyond, without any downtime. In addition, metadata is customisable, allowing more specific information for each piece of content.
Unlocking value through metadata
The metadata tags are a key advantage with object storage — they enable much better identification and classification of data. You can think of objects as being self-describing: they have descriptive labels assigned by the user or application that writes the object. Through a search function, users can easily look for a specific object, even if the data itself is not easily identifiable (such as an image, media clip or data set).
How object storage handles metadata is critical as it drives so many functions of a modern media organisation, from automation and analytics to SEO. Metadata can help business leaders and editors know what content performs well. A strong metadata system can also power centralised searches, helping find assets across the storage infrastructure. After all, you can only monetise and repurpose content that you know exists and can easily find.
4K, 8K, AI and beyond
As new, higher resolution formats and AI tools enter the workflow, capacity requirements will only increase. High capacity solutions are needed, and they exist. It is possible to store 1.5PB of data in just 4U, or 7” of rack height, and 55,000 hours of 4K video content (UAVC-4K, Ultra HD format) in just 12U of rack height, representing a 75% space savings vs an LTO-8 tape library with the same capacity, as examples.
Media organisations should look for automated data protection, replication across facilities for collaboration and disaster recovery, and rich metadata tagging that enables Google-like search and facilitates AI/ML applications, among other things.
In short, object storage can be your best ally in the struggle to manage, protect and leverage your media assets.