Streaming is progressively replacing broadcast as the primary form of video content distribution, driven in large part by the success of global over-the-top (OTT) video platforms such as Netflix, Disney Plus, and Amazon Prime. Grand View Research predicts that the value of the global video streaming market will grow from $59.14 billion in 2021 to $330.51 billion in 2030, at a compound annual growth rate of 21.3% over the forecast period.

A significant technical challenge with streaming is that it implies one-to-one connections from the client to the network servers, whereas broadcast distributes the same video signal for all users (see Figure 1). In other words, 1 million viewers watching a hugely popular sport event requires 1 million physical replicas of the same content on the network, compared with only one for broadcast.

Figure 1. A comparison of broadcast and streaming distribution

As a consequence, video traffic has been growing exponentially. The popularity of streaming creates additional load on networks and requires the deployment of new network infrastructure, increasing energy consumption.

Strategies for Mitigating the Energy Expenditure of Video Traffic Growth

Some will argue that increasing hardware performance will mitigate the carbon footprint of video delivery, despite increasing traffic volume. But this assumption is risky, and in any case, increasing hardware performance to reduce carbon outputs is far from reaching the environmental goal set by IPCC and majority of governments. Regulations have been put into place not only to stabilize the carbon footprint but to dramatically decrease it in the coming years. (See Figure 2.)

Figure 2. Global CO₂ and CH₄ emission trends, courtesy of Carbon Monitor

The good news is that the amount of energy used to deliver streaming services can be reduced. There are three key ways to make video delivery more energy efficient while leveraging existing infrastructure:

• First, third parties, including content providers, can use the ISPs’ networks to decrease the need for new infrastructure and move delivery closer to end users (which also improves QoE). Ultimately, collaboration between ISPs and content providers can result in more revenues for ISPs, allowing them to maintain their network and improve their energy efficiency.
• Second, content providers can stream video via multicast ABR. With this approach, only one stream is delivered over the network to address millions of viewers compared with one stream per viewer in a  traditional ABR delivery scenario.
• Third, operators can continuously optimize the integration of the CDN software on hardware, with the objective to reduce power consumption for the same streaming throughput.

Ultimately, relying on incremental optimization of network components won’t be sufficient for what is at stake. Industry professionals need more data and standardized practices to be able to speak the same language, work together, and develop best practices. Clearly, this requires coordination across the ecosystem. If streaming platforms, telcos, and their technology providers are committed to making an environmental impact, they need to work together to find new ways to deliver video.

How Content Providers and ISPs Can Collaborate Via Edge Caching

A good starting point of collaboration between content providers and ISPs is to minimize the deduplication effect mentioned previously and ensure that the same program is never unnecessarily replicated continually on networks. This is a problem that ISPs have already addressed with a simple solution that could easily be extended to third-party content: edge caching.

The principle of edge caching is to send content only once and to cache popular content deep in the ISP network so that duplication and streaming are done as close as possible to the end user. This allows for dramatic savings in network infrastructure (see Figure 3), which consequently reduces the environmental impact. Ultimately, it’s a win-win situation for everyone: end users can enjoy better streaming quality, which benefits both the content provider and the ISP. 

Figure 3. Video content delivery pathways with and without edge caching

Moreover, external streaming content making use of ISP caches can, at the same time, take advantage of all the software optimization that ISPs have been implementing to distribute their own content. ISPs have been offering video services long before OTT platforms gained prominence. Content providers could leverage ISPs’ delivery experience to their own advantage, improving environmental sustainability.

One good example of such an optimization is for content providers to leverage the IP multicast capacity of ISPs whenever available. OTT ABR can, thanks to multicast ABR (MABR), transit throughout the network in the same one-to-many mode as broadcast and be delivered as ABR directly in the home network via a dedicated conversion process. The conversion typically takes place in an IP gateway or a set-top box. (See Figure 4.)

Figure 4. Video content delivery pathway with MABR caching

Another example is caching elasticity — namely, the ability for ISPs to progressively push the edge of their network further and dynamically adapt cache instances on the actual streaming demand from end users. This evolution is particularly relevant for mobile networks and has been popularized by the Multi-Access Edge Computing (MEC) initiative developed in the 5G standard.

The concept of ISPs sharing their video distribution with third-party streaming content providers — often referred to as “Open CDN” — may seem technically straightforward, but it can be complex in terms of defining the relationship between the two entities:

• If a content provider has an international offering, it will likely want to work with a few different ISPs to establish a relevant footprint. Conversely, ISPs with such capability will want to onboard many content providers to rationalize their investment. The technical interface between them must be as simple to implement as possible to ensure scalability. Several technical tools are being developed with that purpose in mind. The SVTA Open Caching standard is expected to play an important role in setting up a common technical framework.
• Content providers and ISPs have only recently started discussing business agreements, and it will take time before the topic matures enough to reach a consensus on who provides which service and for what price. That said, content providers today have normalized business for their content delivery with public CDNs, and Open CDN could use the same model as a starting point of reference, potentially accelerating its adoption.

Conclusion

Looking at the current situation, the potential for improvement is huge, and caching all streaming content deep in ISP networks is one of the most obvious approaches to start with. If video streaming stakeholders want to comply with the minimal environmental targets, it is inevitable that they will need to start collaborating.

Of course, collaboration will require content providers and ISPs to develop tools and practices to make their interactions as simple and scalable as possible. The outlook for that happening is optimistic, given that everyone would benefit: the content provider will realize better service quality; the ISP will optimize infrastructure costs; and the planet will be greener since there will be less network equipment deployed.

​