The population of the USA “likes” something on Facebook four million times every minute. If that seems like a staggering amount of engagement, then it might also come as a surprise that the average person in the USA touches their smartphone 2,600 times per day. It seems that people check their phones morning, noon and night and smartphone addiction is a recognized condition. Our love of phones – or, more accurately, of all the data and the apps on them – shows no sign of abating. By 2021, the number of devices connected to IP networks will be more than three times the global population (Cisco). Traffic from mobile and wireless devices will account for more that 63 percent of all IP traffic and mobile data traffic will reach 48.3 exabytes per month.
Welcome to the exabyte age. Your next stop will be zettabyte – and that era is almost upon us. 5G will usher in greater speeds and more data traffic than ever. Traditional Wide Area Network (WAN) infrastructure is already creaking under the pressure. If operators’ current concerns are how to minimize dropped calls or buffering videos – just imagine the size of the challenge when billions of IoT devices are added to the mix.
According to IHS, an analyst outfit, the number of connected IoT devices will surge to 125 billion by 2030. These seriously complex networks, flooded with increasing volumes of data traffic and connected devices, could be the baptism of fire for operators relying on the old way of doing things.
Software is re-defining the WAN
It is becoming impossible for humans to efficiently analyze these huge data volumes. Already, there is a pressing need for a smarter, stronger network core that is up to the job of managing this new networking reality. When it comes to re-booting the WAN, it could be that Artificial Intelligence (AI) has the answer. The saying, “software is eating the world”, is certainly true for networking. Software Define Networking (SDN) has been around for around ten years, but not until recently was it possible to deploy these principles in the WAN to make it SD-WAN.
By integrating AI and machine learning (ML), SD-WANs become stronger, smarter and faster. This new breed of WANs, known as autonomous WANs, have self-learning capabilities to add to their lightning-fast data processing abilities. The SD-WAN era will finally see zero human touch and intelligent feedback loops creating value for both the operator and the customer.
Everyone wins with SD-WANs
Automating the network using AI and ML does not just increase speed, improve employee productivity and boost operational efficiency – although all these are extremely valuable benefits. It also results in a consistent, error-free network. This is great news for customers and operators alike. During recent trials of SD-WAN infrastructure and services, customers have experienced higher levels of service, satisfaction and were less likely to churn.
Autonomous WANs have demonstrated that they are able to pre-emptively and dynamically detect network issues, often before users were affected. Using an autonomous WAN solution means that operators are assured of guaranteed SLAs. In turn, this means avoiding costly SLA penalties and improving the brand and reputation.
Intelligent security (for Ariane Grande and Drake too!)
Mobile networks are often the front line of cyber warfare. Operators need to be ever-vigilant for nefarious network activity. One of the best ways is to constantly monitor for unusual traffic patterns. However, it is not always easy to distinguish a DDoS attack from a traffic spike due to, say, the latest album release by Ariana Grande or Drake.
Machine learning algorithms can interpret vast amounts of traffic behavior and an autonomous-WAN can predict performance issues before users are affected. By cross-referencing with algorithms that scrape social media feeds, autonomous-WANs can very quickly confirm whether the network is truly under attack or if millions of excited “Beliebers” or “Arianators” are clamoring to download the latest release.
Next stop: end-to-end automation
The future of networking is autonomous, virtual, agile….and self-learning. By combining AI models, semantic telemetry and intent-based orchestration capabilities, the next-generation of mobile networks are designed to seamlessly manage the growing influx of data through carrier-grade end-to-end automation. Through its capacity to analyze, learn and heal, autonomous WAN networking results in faster troubleshooting and quicker issue resolution.
Traditional tools simply do not have the power to analyze heterogeneous data at ultra-high speeds. Mobile operators can ill-afford network downtime or poor QoE. At best, frustrated subscribers will sound off on social media. At worst, they churn in droves. With the arrival of 5G, today’s data economy will only increase further. There is a pressing need to update and strengthen the core network. Autonomous-WANs address these challenges with automation – they can self-configure, self-manage, self-heal and self-protect with zero human intervention. Welcome to the future of networking.
The post Strengthening the core—how AI is improving mobile networking (Reader Forum) appeared first on RCR Wireless News.