By Naveen Narayanan
Cloud-native or cloud-compatible?
The difference between OTT architectures that are built to maximize the advantages of the cloud and those that leverage peripheral cloud elements is not always obvious, but when you know what to look for, it’s easier to separate the contenders from the pretenders.
Before we explore the differentiating capabilities, however, let’s backtrack to a discussion of why architecture matters. As OTT has evolved into the primary video source for a majority of the television audience, three factors – technology evolution, consumer demand and competition – have driven the need for rapid evolution of service offerings.
Over the past year, services built on legacy platforms have seen the need for flexible infrastructure that could more rapidly and cost-effectively accommodate shifts in the OTT environment. Rising usage and missed opportunities for new features and monetization have shown how existing architectures that only use cloud on paper can’t stand up against rapid growth.
So what separates the cloud-native wheat from the cloud-compatible chaff? At its simplest, it comes down to:
Performance and Cloud Economics – Cloud-native solutions have an advantage in terms of performance when compared to other solutions that leverage a few peripheral cloud elements such as containers and virtual machines . We’ve seen a 10x performance in terms of start and response times for microservices written in GO when compared to traditional Java services deployed in containers. Furthermore, cloud-native solutions leverage stateless architectures that contribute to a highly performant (responsive) API layer, in addition to enabling: scaling to meet demand as viewership increases; load balancing to accommodate high usage in specific areas; repairs to expedite correction of technical issues; and rollbacks to earlier software versions to alleviate service problems.
A cloud-native solution drives significant reduction in cloud spending when compared to cloud-compatible solutions. We’ve noticed a 40% reduction in cloud spend when using on-demand resource management and scaling techniques.
Automation – Cloud-native architectures leverage automation in three key areas: infrastructure deployment for rapid time to market; instantaneous resource allocation for efficient scaling; and DevOps – in our case, Continuous Integration/Continuous Delivery (CI/CD) supporting Canary deployments and replacing traditional Waterfall models to help unlock business value quickly. At Quickplay, our Gen5 multi-tenant platform is designed to utilize automation in each of the key areas mentioned above. This enables the solution to be agile while supporting complex media workflows and allows us to launch new services in weeks and new capabilities for existing services in days.
Modularity – In a cloud-native architecture, capabilities are implemented as a series of atomic functions that come together to build the final end product. Our microservice architecture allows OTT providers to build on top of extensible modules and focus on innovation. Modularity enables change control for services, which can be forked to serve different use cases and won’t impact the entire system if one of the microservices misbehaves. Lastly, modularity is directly related to the maintainability of the codebase. Maintenance costs can be reduced by as much as 40% when development occurs on a modular architecture with an automated CI/CD pipeline.
Zero-Trust Security and Scalability – To maximize deployment flexibility, every module in a cloud-native stack – all the way down to the granular microservices – should be designed to eliminate the possibility of breaches, whether it is deployed in a public or private cloud, multi-cloud environment, on-premises, or in a hybrid fashion. We ensure that any authorized viewer on any device can access our customers’ services; at the same time, our Quickplay team improves on traditional perimeter security paradigms by ensuring that each component has its own independent authentication and trust built in. The above security model enables us to scale quickly across different infrastructure targets in a secure fashion while providing full observability to our support teams through a service mesh interface.
Innovation Readiness – The agility of cloud-native platforms enables constant evolution, including the addition of business logic, the creation of new microservices, and the integration of third-party solutions. In our case, these extensions and integrations can be done by the service provider or supported by our team. We provide a managed service solution to help our customers keep with the pace of technology in the media space and experiment with new features that drive user engagement and increase revenues and/or decrease costs for the service provider.
The brief history of OTT has been one of continuous transformation. As the industry has moved from early cable or broadcast extensions to today’s services that are destinations in themselves, the ability and willingness to embrace new technologies that increase effectiveness and reduce costs has separated winners from losers.
The shift from legacy cloud-adapted to full-featured cloud-native technologies presents the industry with another quantum-leap moment that will help to differentiate forward-looking services.
Does your architecture have what it takes to make that jump?