Most people are well aware that a streaming-based architecture was originally used for the streaming of video content. That is no longer the case, however. Today, these systems are used in millions of interactions per day with various digital systems. Tens of thousands of services and websites, that are not video services per se, use a streaming architecture. From online banking to social media to real-time online gaming — users have a set of expectations about how quickly systems should respond and how few moments of downtime should be allowed before it negatively impacts their experience. These expectations are new, they are different, and the architectures that have evolved over years to support these use cases are very different from those that have traditionally been known and understood.
From passive viewing to active interaction
Streaming-based systems are now applied across a wide range of digital platforms. A notable example can be found in real-time entertainment environments, where continuous interaction is essential. Services such as live casino platforms demonstrate how streaming infrastructure supports immediate engagement by combining live video delivery with responsive user interfaces. Recently Facebook and Microsoft announced that they are shifting their content centre of gravity to Europe from North America and Canada due to global shifts in how people consume digital content. The Reuters Institute Digital News Report highlighted a growing global expectation for more real time information in more places, and that technology platforms have to deliver lower latency and more dynamic content delivery technologies to reflect this new reality.
The architecture driving real-time experiences
What makes a video stream look so smooth? The IEEE Special Report on Research in Streaming Media addresses the technological issues involved in communicating over streaming media protocols. Issues such as packet arrangement, buffer management and link quality have to be resolved in real-time. It’s a challenging task; one that must accommodate the physical limitations of current-day networks while providing an enjoyable viewer experience – even as millions of viewers simultaneously access the same stream. Platform engineers are now engaged in a lot of work on edge computing, and making the most out of Content Delivery Networks to improve latency at scale. It’s a very intense engineering effort and the architecture of a streaming system continues to evolve to address these needs.
Security and reliability remain critical concerns
New Internet connections, new Internet threats. ENISA’s Threat Landscape for Online Platforms report identifies numerous threats to these streaming platforms, ranging from distributed denial-of-service attacks, data interception to session hijacking. To combat these risks, these threats must be built-in into the infrastructure and not an “add-on” for CSPs.
To ensure a safe digital engagement experience, it is vital that robust authentication is in place, that data is encrypted in transit and that real-time monitoring is ongoing. While technical challenges can occasionally occur on real-time engagement platforms, any platform down time has a profound effect on users and must therefore be kept to a minimum.
A structural change, not a passing trend
This shift is likely to be permanent as a result of trends in global streaming habits. As audience behaviour continues to evolve and demand for real time streaming and more engaging experiences is expected to rise, streaming platforms across industries will have to invest significantly in infrastructure and cybersecurity. The benchmark for present day platforms, such as YouTube, Facebook and Netflix, will likely set the bar for consumer experience expectations for several years to come.
