The Internet is Today’s Real Time Network

Today, nearly 3.4 billion people, 46% of the world’s population, are connected to the Internet (InternetStats) and enterprises of all sizes rely increasingly on the Internet, as well as their private IP networks, for every moment of their business activity. Interestingly, according to Ericsson’s latest Mobility Report, there are also 3.4 billion smartphones in the world that will grow to 6.4 billion over the next six years.

Now, it is not quite the case that everyone is accessing the Internet only through a mobile device, but this is becoming closer and closer to the truth. For a huge and growing number of global users their mobile device is their primary screen. Social media platforms like Facebook and Twitter now report many more mobile interactions than desktop interactions and most new social media apps, such as Instagram, WhatsApp, WeChat, and Snapchat, are primarily mobile.

Our mobile smartphones have become our communications hubs. Over just the past few years, real-time communications has completely changed for global consumers. For example, voice and video communications capabilities are now fully embedded in Apple phones with FaceTime, Google Hangouts, Facebook Messenger, WhatsApp, Snapchat, WeChat, Line, Vibe, SOMA, Xiaomi phones and more – while classic Skype runs 3 billion minutes a day for 300 million users. As a result, there are now billions of consumers worldwide who expect easy global contextual click-to-call and click-to-video, right in their everyday apps. Ask the kids – millennials and centennials who wouldn’t dream of making a traditional phone-call now expect their smartphones and apps to instantly connect together all their friends from around the world. This is a huge generational change which will have dramatic market impact on traditional networks and telco approaches.

And this is just the beginning. The broad availability of cloud “communications platform as a service” (CPaaS) choices, open WebRTC capabilities, and other real-time technologies, have opened real-time voice and video to all software developers for all kinds of application. Whatever mix of technologies developers choose, there has been a major industry attitude change – every business or consumer app can have fully integrated communications and any developer can easily add this without being a real-time expert.

Meanwhile, Inside the Enterprise

But what about the enterprise? Running voice communications over IP networks (“VoIP”) got its start in 1995, but it has taken several decades for enterprises to build out their IP networks with the bandwidth and QoS (quality of service) to make them comfortable moving their enterprise voice and then video to IP. However, the result is that for enterprises this has meant increased internal network expense. Today, almost all enterprise voice and video requires well-tuned private networks and expensive MPLS-type telco connections between locations and countries.

Can the Internet provide the voice and video quality needed for business communications?

For global and mobile real-time (and real-world) solutions to continue their rapid growth, the answer has to be YES.

This means the key challenge for the industry is ensuring global quality of experience for voice and video over the Internet, taking account of challenging and varying network conditions and devices. However, traditional approaches are often “upside down” with a built-in assumption that the network is separate layer that must be corrected before voice and video can be used. But this is not the right way up for a real-world Internet future. In the wild Internet, without built-in QoS, the communications layer itself must adapt, rapidly and in real-time, to wildly varying network conditions.

Fixing the Internet for Real-time Video and Voice

To achieve reliable global communications over the Internet with high quality requires new network innovation “in the middle”, at three levels:

  1. Internet Backbone Bottlenecks – globally there are a lot of Internet bottlenecks that can be time-of-day delays or just appear and disappear according to global traffic, especially in and out of certain countries. To “add back” QoS for the Internet, new solutions must route around these bottlenecks based on a global understanding of the real-time conditions of different Internet network paths.
  2. Mobile Device Issues – not everyone in the world has the latest iPhones or top of the line Android phone, especially global customers. Initial quality issues must be tackled right on the device, including video encoding decisions based on CPU/battery, echo cancelation, noise removal, active audio detection and other technologies, appropriately tuned for different mobile devices with differing performance.
  3. The Mobile Last Mile – the final link to mobile devices can be a big problem. These devices are, by definition, mobile and move in and out of good coverage and move between 3G, 4G and Wi-Fi connections with varying bandwidth and performance. Along with an intense focus on error correction and concealment, solutions ideally need local points-of-presence to optimize dynamic response and packet re-transmission when possible. Pure “peer-to-peer” solutions will typically be less effective for mobile-to-mobile communications versus an adaptive network.

Conclusion – Choose Wisely

Real-time communications is expanding rapidly with video and voice becoming embedded in nearly every application, and the Internet is becoming the back-end real-time network for these new applications. Application developers therefore have critical new choices to make – more important than previous codec and client library choices, developers need to choose network platforms that deliver the best end-to-end quality and reliability over the Internet. Choose wisely!

 

Get started with the Quality-of-Experience (QoE) real-time network offered by Agora.io from here.

* This article first appeared in the July 2016 issue of TMC’s Internet Telephony Magazine.

Image Credit / Flicker User: Alex McCabe