The demand for data will continue to explode, and of that, there is no doubt. An IDC report puts the amount of digital data in the world in 2025 at 163 Zetabytes (ZB). Of course, all of this data is not going to be stored long term, but 11.1 ZB is predicted to be stored long term by 2025, and 60% to 80% of this data is expected to be cold or archival data. Much of this is driven by Sports, Media and Entertainment (but also Oil & Gas, Healthcare and Surveillance), and those content owners have ever-growing lakes of valuable content that they are either monetising already or are planning to do so. This is going to be made possible, in large part, by a more widespread adoption of Artificial Intelligence (AI) and Machine Learning (ML) tools to make sense of the vast volume of data.
One of the biggest trends at the moment is processes relating to optimising where content and assets are stored. There seems to be a lot of truth in the “cloud boomerang” a phenomenon in the Sports, Media and Entertainment industry in which companies are egressing content from public cloud providers and storing, once more, on-premises. Although this trend is being driven primarily by financial concerns, over both recurring and escalating cloud storage costs, there are other advantages of this migration out of the cloud for customers.
Companies that are rich in content usually have their media distributed over multiple geographies. This is where cloud native solutions add a tremendous amount of value, with the ability to easily search metadata and deliver content on demand. In contrast, many public cloud providers do an excellent job of actually siloing content in geos making it difficult to share content within an organisation. To pull together customers’ public and private cloud storage, Media Asset Managers (MAMs) are going to be increasingly leveraged, not only to index the content contained within them, but also to build or enhance workflows, and provide a single point of access across an organisation.
AI and ML applications that provide transcription, object recognition, and transcoding tools, for example, can be easily set to work on these newly created content lakes. Traditionally many of these tasks were carried out by teams of media professionals, interacting with MAMs, and painstakingly inputting or logging data. However, the ability to bring AI and ML applications to the content, where performance and processing are unaffected by the constraints of an organization’s connectivity to the public cloud, or the need to migrate between public cloud vendors to access the desired applications. Likewise, content can also easily be accessed by OTT applications to provide an immediate pathway for many companies to make it easy to monetize their assets.
Choosing a cloud-native solution such as SymplyPERIFERY to host on-prem workloads, provides a rock-solid platform on which to consolidate valuable content, in order to maximize its longevity, governance, security, and sovereignty while applying AI/ML toolchains to harvest the content’s inherent value to an organization. What’s more the performance and scalability of SymplyPERIFERY remains under the direct control of customers, who not only benefit from improved performance, but also from consolidation, and the ability to more accurately predict their costs.
As a company, Symply helps customers implement this on-premises-first (OPF) Hybrid Cloud infrastructure model, which prioritises on-premises workflows enabling customers to unlock the advantages already discussed and treats the public cloud as an enhancer that can be removed or changed without disruption to operations. Thus, enabling organisations to preserve operations, applications, function, and performance, while allowing them to continue to grow and protecting them against the cost unpredictability of the public cloud.
