By Russ Fordyce
Senior Director, Content Marketing
Trends from the Streaming Media East conference are pushing network operators to deploy more flexible and scalable network architectures
More than 1,000 people just left the Streaming Media East conference in New York City. With over 100 speakers, six workshop tracks, and more than 40 solutions on display, the streaming media (over the internet) movement is on the rise.
According to eMarketer, there are now more than 33 million cord cutters in the United States, and that group is expected to grow to 55 million by 2022. Furthermore, a CNBC study shows that 57% of Americans now have some form of streaming service, and even more interestingly, nearly all Americans ages 25-34 access TV content through the internet.
So, internet streaming is here to stay and growing. But what’s new and what will impact network operators in the future?
Real-time Latency Is a Must
Streaming media providers are still struggling to get to parity with broadcast and cable TV providers in terms of customer experience. Those of us who remember watching actual TVs know that pushing the channel up or down button instantly brings a whole new program into view.
But that’s not true with virtually all streaming platforms. Want to watch a different show on Netflix, Amazon Prime, or BritBox? Push the back button, scroll left or right, or maybe up or down. to find a show. Once you find one, click and watch the “wheel of waiting” as the program is pulled into your platform of choice and buffers for a few seconds before starting.
Netflix has done a good job at improving customer experience by offering innovative solutions like streaming previews and starting streams at lower video quality until the network and cache catch up. While this makes the experience better, it doesn’t give us the instant gratification we all desire!
New media standards, protocols, and techniques may help reduce latency as well. John Gainfort, Development Manager from RealEyes Media, detailed in his presentation how Chunked CMAF (Common Media Application Format) encoding, an emerging standard intended to simplify delivery of HTTP-based streaming media, can help reduce latency in certain deployments.
This process involves breaking the video into smaller chunks of a set duration, which can then be immediately published upon encoding. That way, near-real-time delivery can take place while later chunks are still processing. Sounds like live video is getting the old TCP-IP treatment to me!
webRTC Adoption for Interactive Streams
Web Real-Time Communications (WebRTC) was also a hot topic as providers look to this protocol to improve customer experience. This technology offers real-time interactivity without a plugin. Instead it relies on technology and standards already baked into most browsers and some smart TVs.
The industry is still unsure if “broadcast quality” is a real goal for WebRTC. But the streaming media industry doesn’t stop at TV-like content. Many providers are looking for new content that is interactive and collaborative: think game shows, auctions, and, of course, online gaming and gambling.
In a recent article, Traci Ruether from streaming media software provider Wowza said, “from video meetings to job interviews, live streaming with WebRTC lends itself well to today’s professional settings. Other real-time streaming scenarios that could implement WebRTC include interactive game apps designed for small groups.”
When you add into the mix the rapid adoption and deployment of 5G networks, WebRTC could be the next protocol to watch as streaming media providers push new interactive services into consumers’ homes and offices.
More and More Personalization
The “holy grail” of streaming media is personalization – making content more tailored to the individual watching. In theory, the more personal the content, the more you’ll watch. This is good for subscription and ad-supported business models.
Netflix and others have spent resources tweaking their platforms to recommend content viewers may like based on other factors. In fact, at one point Netflix even went as far as offering the Progress Prize for anyone who could improve its recommendation algorithm by 10%. But at the Streaming Media East conference, the talk was not on programming but instead on advertising. Streaming media providers have access to more data points about their consumers than broadcast TV providers or even cable operators.
Over the top (OTT) providers are looking to create dynamic ad insertions by user, by device. So instead of the broadcast model, with the same ad for everyone, or the cable TV multicast model of national and regional advertising, streaming media could send individual streams with unique ad content for each stream or user.
Scott Apgar, Senior Director, Advanced Advertising for Seachange, noted in his presentation that this type of dynamic ad insertion will take some time to gain adoption as the cost per ad would go up, the back office software platforms needed to house and deploy content are not mature, and the equipment to make it all happen is expensive.
Challenges Streaming Media Providers Can’t Solve on Their Own
Most streaming media business models are independent of broadband providers and rely on last-mile connections from cable and telco operators. And, as noted above, latency remains a major issue, especially as interactive and collaborative content and services are developed.
To really solve for latency, broadband operators must deploy computing and content resources closer to the consumer. Many operators are looking to multi-access edge computing (MEC) architectures that package and integrate servers with routing, and sometimes optical network equipment, so that they can be deployed in neighborhoods, cell sites, or local data centers.
Network operators themselves will also need to scale metro networks quickly as services gain adoption. At the edge of the network, scaling up to support these streams will require a new architecture so that costs do not scale with demand and so that the edge is more programmable and automated. These new architectures will rely on disaggregated router and switch platforms instead of costly chassis models. This new class of routers comprises application-optimized, carrier-class white boxes designed to enable networks to scale quickly through open interfaces and application programming interfaces.
With new network resources come new network complexities. From the IP edge to optical core, network automation and ultimately artificial intelligence will be necessary to keep up with all the changes your new network will require. Software platforms already exist to reduce complexity and improve customer satisfaction by creating a network that thinks for itself.