Home Technical Articles Network Performance and Optimizations are Essential for Viewership Scale
Applications

Network Performance and Optimizations are Essential for Viewership Scale

About The Author

Outline

This blog is the third in a three-part series on achieving operational scale for your streaming business. Read part one here and part two here.

The video streaming ecosystem continues to expand with new devices and platforms. Streaming technology must be flexible and powerful enough to accommodate the endless operating systems, devices, endpoints, and business rules. In part two of the series, we used the complexities of supporting digital rights management (DRM) across many devices as an example of playback scale. Many other business requirements, including DRM, must work together for a seamless viewer experience.

Ultimately, your viewers should be focused on content rather than on the quality of the stream. You need streaming technology that can keep up with the industry, provide support for a broad range of platforms and devices (each with its nuanced workflows and requirements), and reliably manage viewing sessions at scale. But your network provider also plays a pivotal role in reaching all your viewers.

Your streaming technology is only as good as your network

Capacity, storage, and reach are the backbone of the networking world, and your provider must have plenty of each to deliver video effectively. Reliable and reputable providers will offer quality-of-service (QoS) guarantees, flexible quality-of-experience (QoE) refinement tools, and 24×7 monitoring and NOC services to stay ahead of potential service interruptions.

Edgio’s delivery network offers 250+ Tbps capacity across 300+ global PoPs with 7,000+ interconnections with extensive caching power and a Super PoP architecture designed to support the largest surge in video demand. Uplynk is our industry-leading streaming technology integrated with the Edgio network. Keep reading to understand how we’ve fine-tuned our network for optimal media delivery.

Using AI to manage delivery performance

Delivering video streams requires tools and instruments to unwind the incredibly complex interconnections and providers between your content and your viewer. With more than 90,000 global ASNs, your network should simplify these complexities with established peering points and a proven strategy for choosing optimal paths for routing video from the server to the client. As a CDN provider, the complexity involved with connecting to thousands of ASNs is compounded by the need to accommodate each customer’s performance requirements, usage profile, type of traffic, and many other factors.

Our goal at Edgio is to provide our customers with predictive adjustments that outpace the need for reactive actions. We employ artificial intelligence to help us process the vast amounts of data we gather about the performance of our network and your service to identify the real-time actions we can take to tune network performance and optimize delivery. Read our technical article on the optimizations we made for a major North American vMVPD streaming service using robust tools that continually monitor round-trip times (RTTs), including internally-developed tools like Stargazer to isolate problems or identify network routes or configurations that optimize performance.

Network optimizations for live streaming

CDN responsiveness and speed are vital for live video infrastructure. Without it, videos are slow to load, streaming sessions stall and buffer, and viewers end up frustrated and unhappy. Delivering a smooth stream to thousands or even millions requires low-latency techniques and special caching configurations.

Reducing latency has increasingly become a focus for streaming service providers, and segment sizes play a significant role because the player has to download a preset number of segments before it can start playing the live stream. This allows the player to buffer enough video to ensure smooth video playback without rebuffering when there is congestion in the network. However, this also puts the stream behind live from the outset. A shorter segment size may seem like a fast and easy way to reduce latency, but the impact is significant, from additional overhead to increased risk of rebuffering.

While reduced latency is only part of the equation, how well the CDN can interact with the origin server and the clients while scaling for large viewing audiences that come all at once is especially important in live streaming. The CDN must be highly efficient and intelligent about how it fills the cache and how it handles client requests during and, even more importantly, before initiating the cache fill process. The CDN should be able to keep the load on the origin server to an absolute minimum while avoiding adding too much extra latency to the entire media pipeline. This ensures that viewers enjoy smooth and continuous playback. We offer robust caching optimization features that maximize origin offload and improve the end-user experience, like Origin Shield, Partial Cache Sharing, and Dynamic Hot Filing. Read our technical articles to learn more about low latency and our cache optimizations for live streams.

Multi-CDN strategies to meet performance and/or reliability objectives

Multi-CDN strategies ensure viewers experience a high-quality, always-on experience regardless of location or device. For instance, depending on your traffic profile and footprint, you may benefit from implementing multiple CDNs by routing traffic to a particular network that is known to perform well for that region. You can also use a multi-CDN strategy to avoid downtime during maintenance or an outage, provide faster page load times and serve more objects from cache, minimize damage from cyberattacks if a CDN becomes compromised, and reduce bandwidth costs by balancing out loads to reduce bandwidth costs.

But not all multi-CDN strategies are the same. If you’re satisfied with your current CDN’s performance and your main objective is to improve reliability and prevent downtime, your plan will look completely different than someone wanting to improve performance, which requires a monitoring strategy. At Edgio, we have tools that collect network congestion data and automate where requests land on the edge. These automated systems look at the metrics we collect, provide real-time feedback about those metrics, and make continual iterative decisions about where traffic should land. Whether your goals are better performance, better reliability, or a combination of the two, be sure your network providers offer the compatibility and flexibility to configure traffic to meet your goals.

Edgio’s delivery network was built for media

Edgio has one of the world’s largest networks with 250+ Tbps of capacity, 300+ points of presence (PoPs) across six continents, and more than 7,000 interconnections with last-mile networks. Our advanced architecture rapidly serves up content with immense computational power, IP Anycast routing, and Layer 7 load balancing to reduce latency and serve more assets from the cache. We power billions of daily digital experiences, from social media to banking, to game downloads, to watching your favorite shows and movies online anywhere in the world, with the technology, services, and support to power businesses 24 hours a day, 365 days a year. Our streaming product, Uplynk, is tightly integrated with the Edgio network, and with a near-perfect uptime (99.9%), we guarantee your viewers have the fastest and highest quality experiences.