This week, telecommunications experts will gather in Barcelona for Mobile World Congress, the industry’s premier event and a chance for tech companies to show off their latest innovations in areas as diverse as 5G technology, artificial intelligence, and the cloud.
Although the cloud is far from a new idea, its true capabilities are only now beginning to be realized. Here are four ways in which the cloud will shape our lives over the next decade and beyond.
Building Digital-First InfrastructureCloud will provide the digital infrastructure of tomorrow’s cities, where an estimated 6 billion of the world’s population will live by 2045. Smart elevators and parking lots, driverless cars and drone taxis, trains and subways, farms and power plants -- all will be safer and better managed, thanks to the cloud’s ability to store and analyze data.
The cloud will also be transformative for companies, especially small and mid-sized businesses, as data analytics, artificial intelligence and other capabilities become available as services. Because each industry has different needs, Huawei, a global tech company where I head up the communications team, is working on what we call the Industry Cloud: thousands of distinct, separate clouds, all working in concert across a digital ecosystem of different industry verticals. For example:
- A commercial aviation cloud will help airlines manage ground operations such as maintenance, fueling, baggage handling, and cabin cleaning, thereby increasing efficiency and helping flights take off on time.
- A utilities cloud will automatically repair faults in the power grid to ensure that homes and businesses get the electricity they need.
- A banking cloud will let financial institutions scan thousands of transactions per second to prevent fraud.
Regardless of industry or size, all companies need digital infrastructure to support their business operations. But cloud will change ICT from a support system into a production system. For example, OpenDesk, a London-based company, uploads furniture designs to the cloud and lets customers download the designs and manufacture the furniture locally. This lowers shipping and inventory costs, while reducing the company’s carbon footprint.
Managing DataThe cloud will also help society cope with growing volumes of data. This includes applications like high-definition video, which Huawei estimates will account for 89% of individual user traffic by 2025.
Soon, network bandwidth and storage requirements will be driven less by user-generated cat videos and more by what IDC’s Data Age 2025 white paper calls “image and video content for non-entertainment purposes.” Examples include advertising and video recordings used in public safety applications. IDC also cites “productivity-driven” data such as files on PCs and servers; meta-data about digital files and web pages; and data created by machine-to-machine communications in the Internet of Things. Cloud will enable us to store this rising tide of data and mine it for usable insights.
Artificial IntelligenceThe cloud will support emerging technologies such as artificial intelligence and help them to adapt to new platforms such as mobile. When smartphone sales overtook sales of PCs in 2011, mobile became the world’s largest computing platform. Naturally, AI has made its way into phones.
But smartphones capture lots of unstructured data such as emails, text messages and photos. Analyzing unstructured data takes time and processing power that most smartphones don’t have, so the phones have to send the data to powerful servers in the cloud. This slows down AI’s response time.
The solution: divide the labor between the cloud and the phone. Training, the process whereby AI learns, happens in the cloud. When Google’s AI algorithm plays millions of games of Go with itself until it becomes a grand master capable of defeating a human, it does so in the cloud, where processing power is abundant and AI learns quickly. Inference, where the AI takes what it has learned and applies it to real-life problems, happens on the device.
Smartphones will use inference continuously, according to research sponsored by Huawei. This always-on intelligence will enable the device to respond immediately to voice commands; ensure that photos are cataloged according to content; and set cameras perfectly for different subjects under various shooting conditions.
Because inference needs to process data in real time, all the time, even tomorrow’s super-advanced smartphones won’t be able to meet the computing demands imposed by AI. They’ll have to rely on the processing power of the cloud.
