Cisco: What’s coming up in 2019?
Innovation, Thought Leadership, Networking, SDN, Video
Now is a hugely exciting time to be working in the technology and communications space. As we move into 2019, it’s useful to take a view on some of the upcoming tech trends that will affect businesses.
As predicted, 2018 saw businesses keen to better-utilise the wealth of data within their organisations. Optimising and managing multicloud environments was a priority for many, as was the need to embed security at a foundational level. None of these trends are likely to go away over the coming year.
Otherwise, 2019 is likely to see an interesting blend of ‘evolution’ – the expansion and development of trends already underway – and ‘revolution’, with changing demands forcing organisations to embrace digitization and new technologies at a higher speed.
1. AI-ML will become ‘everyday’
AI and ML have progressed rapidly over the past few years, driven by a vibrant research community, as well as the availability of ML-ready datasets, increases in compute power, and mathematical advances in the field.
The innovations in AI-ML solved long-standing problems, leading to the belief that these technologies can solve many challenges. High-profile consumer applications such as driverless cars and ‘human like’ digital robots have captured the public imagination. This hype is however only partially justified, as enterprises are learning what it takes to adopt AI-ML.
For one, to be able to apply AI-ML, companies must have their digital house in order: A data acquisition and curation process must be in place, connected to proper computing, which feeds systems of summarization for insight visualization. From there, human specialists and experts can make the last mile of reasoning about the implications and the bigger picture. Despite the hype, 2019 will still be a year in which AI-ML will help humans make better decisions and work faster, rather than take decisions on its own. For example, data-centric hedge funds already rely on AI to support new trading models. And in a time of acute shortage of talent, HR departments are looking to AI to enhance talent acquisition and retention.
Closer to home, Cisco has been using AI and ML to solve real problems in a pragmatic way. For instance, we analyse huge amounts of network data, identify and contain threats, or ensure smoother workflows. Cisco DNA Analytics can pull anonymised, ‘opt in’ data from customers’ network telemetry, identify patterns, and apply insights. This allows reductions in the cost of network operation and increases security. Cisco Encrypted Traffic Analytics (ETA) uses machine learning to find malware inside encrypted traffic without decrypting it – an industry first. Moreover, Cisco WebEx Assistant uses natural language processing to understand voice, making ‘dialling in’ to meetings effectively zero-touch. Expect to see further developments in AI-ML happening thick and fast across 2019.
2. The Internet is expanding in ways we could not have imagined
According to the latest Cisco Visual Networking Index (VNI), by 2022, more IP traffic will cross global networks than in all prior ‘internet years’ (up to the end of 2016). In other words, more traffic will be created in 2022 than in the 32 years since the internet started.
Traffic demand will continue to grow at a consistent pace, with global IP traffic growing 3-fold from 2017 to 2022, a compound annual growth rate of 26%.
By 2022, 60 percent of the global population will be internet users, but new users will be met by a completely different internet – one which connects humans in their daily lives through wearables, home appliances and automated vehicles. Less visual, but no less impactful, will be the vast array of IoT sensors – used, for example, in harbours to optimise and secure shipping, or in smart city parking sensors that will help reduce environmental impacts. More than 28 billion devices and connections will be online, and more than 50% of these will be machine-to-machine.
All of this additional human and M2M connectivity will drive a steep increase in broadband, Wi-Fi and mobile speeds: average global fixed broadband and Wi-Fi connection speeds will double by 2022, whilst mobile connection speeds will more than triple, compared with the 2017 baseline.
3. From ultra-HD, to live video and VR, growth in visual media will drive dramatic bandwidth demands
By 2022, video will make up 82 percent of all IP traffic. Within that, live video will grow 15-fold from 2017 to 2022 and will account for 17% of all Internet video traffic, whilst gradually replacing traditional broadcast viewing hours.
Virtual and augmented reality traffic will also skyrocket as more consumers and businesses use these technologies. With new hardware available to individuals, and a growing body of content to consume, VR and AR are expected to continue a high growth trajectory. Traffic associated with virtual and augmented reality applications is poised to grow 12-fold over the next five years. The next year should see more application of VR in business, from demos and virtual ‘test drives’, to assessments of products and real estate online.
4. Mobile connectivity will continue to grow, with 5G moving further off the starting blocks
Mobile data traffic will continue to grow relative to other forms of traffic across 2019. With the move to 5G still in its early stages, 3G and 4G will continue to be the more widely-used protocols across 2019.
The Cisco VNI report estimates that by 2022, 22 percent of global Internet traffic will come from mobile (cellular) networks (up from 12 percent in 2017). By 2022, about 3 percent of global mobile devices/connections will be 5G-capable (and nearly 12 percent of global mobile traffic will come from 5G).
As expected, mobile carriers from around the world are beginning to introduce trial 5G networks (see 5G Availability Around the World from Lifewire). Many industry experts believe that large-scale 5G deployments will actually begin to take shape in 2020, when mobile spectrum, standards, profitable business plans, and other operational issues are more fully fleshed out.
5. Blockchain will continue to see new and innovative uses beyond the enterprise
Blockchain will continue to expand across areas such as supply chain management, networking, digital identity and currency trading. It is likely that all major cloud service providers will have deployed Blockchain commercially by the end of 2019, and it will have become an essential element in most AI and IoT offerings within the next few years.
Just as interesting is Blockchain’s expanding role as a force for good beyond the enterprise. Amongst its many uses, for example, Blockchain is used to validate that minerals in products are from conflict-free sources, to combat child trafficking, and to buy-and sell energy on independent microgrids.
Cisco believes in the value of partnerships with bodies such as the Trusted IoT Alliance, Hyperledger, and the Blockchain Research Institute as a means of exploring Blockchain’s potential. We expect such partnerships play an increasing role in uncovering innovative use cases over the next few years.
6. Businesses will increasingly need to rethink their network
Across 2019 more and more businesses will have to focus on network transformation in order to continue to deliver premium customer experiences.
Today’s networks were not built to accommodate the networking needs of the next few years. Whether 5G, virtual reality, IoT, or AI/ML, behind all of these growing trends is a digital backbone composed of a multitude of public and private networks. And with an expected one million things coming online every hour by 2020, this infrastructure is being placed under increasing strain. Increases in device numbers, bandwidth requirements, and attack surfaces, also mean that manual administration of networks is becoming impractical.
Businesses now need to be able to enable any device, from anywhere, at any time, across multiple domains. What were previously thought of as independent networks – to be built separately and interconnected – must now be united in a single multi-domain architecture. This architecture will need to be highly-automated, anticipating actions and intent to self-optimise, learn and self-heal, and will treat security as foundational.
(CC0 Creative Commons image ref)