- The global datasphere, defined as the total data created, captured and replicated globally, is set to explode over the next several years as the digitalisation of global industries takes hold.
- Data volume is getting bigger exponentially, spurred by the following factors: the growing number of data-generating devices including smart phones, wearable electronics and a whole host of gadgets with sensors; the prevalence of Internet of things and cloud computing; the imminent adoption of 5G; and other technological advancements such as computational photography.
Data continues to get bigger
English data scientist and mathematician Clive Humby first coined the catchphrase "data is the new oil" back in 2006, when the global datasphere—total data created, captured and replicated globally—amounted to an estimated 160 exabytes1. Since then, data, which has become one of the most valuable resources in the world, has exploded in volume, growing exponentially to approximately 33,000 exabytes, or 33 zettabytes,2 at the end of 2018".
As we move from an oil-driven era to a data-driven age that is shaped by the rapid digital transformation of global industries (also known as the “Fourth Industrial Revolution”), data is increasingly becoming voluminous, varied and valuable. The global datasphere has expanded and continues to grow at a breakneck speed. In a November 2018 white paper “Data Age 2025”, research firm IDC predicted that the global datasphere could increase from 33 zettabytes in 2018 to 175 zettabytes by 2025 (Chart 1).
Chart 1: Exponential growth of data
Source: Data Age 2025 report, sponsored by Seagate with data from IDC Global Datasphere, November 2018
At the same time, internet protocol (IP) traffic, or the flow of data across the internet, had also grown 87 times from 2005 to 2018, according to estimates by Cisco—which predicts that annual global IP traffic could reach 4.75 zettabytes by 20223. That is a growth of 222 times since 2005 (Chart 2).
Illustration 2: Estimated growth of IP traffic from 2005–2022
One of the most interesting statistics about the increasing volume of data appeared in a 2016 IBM report “10 Key Marketing Trends For 2017”, which outlined that 90% of all worldwide data at that time was created in the preceding two years, when 2.5 quintillion bytes of data was created per day. With data volume around the world still growing at an immense pace, it could be a matter of time before the global datasphere reaches a brontobyte (approximately 1,024 yottabytes) in the coming decade (Chart 3).
Chart 3: Digital universe could reach a brontobyte in the coming decade
Growth drivers of data
So what is causing the volume of all data created, captured and replicated globally to surge in recent years and grow at a rapid pace? To begin with, the number of data-generating devices—including smart phones, tablets, wearable electronics and a whole host of gadgets with sensors—have exploded of late, generating an abundance of datasets. For billions of smartphone users around the world, the bulk of data usage in their handsets is generally centred on photo and video storage, all which is driving the growth of data globally.
Other segments spurring the growth of data include the following: the increasing global internet and mobile usage; the prevalence of the Internet of things (IoT); an increase in virtual reality (VR) and augmented reality (AR) traffic; the imminent adoption of fifth-generation, or 5G, cellular network technology; and hyperscale computing and cloud computing (or internet-based computing) that is backed by data centres (Chart 4).
Chart 4: Wide ranging factors spurring the growth of data
Source: BroadGroup, Ericsson, Synergy Research and Mordor Intelligence
Indeed, current demand for data centre space is hugely underpinned by cloud adoption, enterprise digital transformation, video streaming, e-commerce and data sovereignty regulations. Cloud computing, for instance, permits access to programs and data over the internet instead of computers' hard drive; it facilitates online data storage, which is increasingly popular with individual and corporate users. Essentially, cloud computing has put the power of supercomputers, backed up by data centres, in the hands of users, who could easily run algorithmic-type executions on online platforms that offer scale and solutions.
On top of that, global data growth is set to be further boosted in the near term as many countries push ahead with the rollout of 5G networks. For example, China in early November 2019 rolled out an extensive 5G network to its telecommunication operators. Elsewhere, hardware vendors have been revising up sales expectations in recent quarters on 5G prospects. Taiwan Semiconductor Manufacturing Company, for instance, expects 5G momentum to spur 5G smartphone penetration in 2020 to a percentage in the mid-teens from its previous single-digit estimate. That is a big number relative to industry estimates such as those of IDC.
5G mobile technology offers significantly faster data download and upload speeds, broader and more extensive coverage, and more stable connections. In a nutshell, 5G provides a three-tier step up in terms of data availability. First, consumers will get to enjoy faster download and upload speed on their 5G handsets. Second, 5G technology will enable billions of IoT devices to connect and communicate with one another in a process known as machine-to-machine communication. The third and the most critical data step up involves machine learning and artificial intelligence, both of which could be integrated and enabled on 5G networks (Chart 5).
Currently, we are just at the early stages of 5G technology. The full spectrum of the decade-long 5G evolution will create a powerful growth driver for the global datasphere as it is going to take a huge amount of data to power it. With the 5G market just starting to grow beyond the consumer experience towards the industrial/machine experience, data usage is going to explode on a global scale.
Chart 5: 5G new capabilities beyond broadband
Source: Credit Suisse
Lastly, technological advancements, in such areas as computational imaging, are also set to boost the rate of data created, captured and replicated in the coming years. Computational photography or imaging takes all the related pictures that are available within the clouds and other sources, and sews them together into a combined image. This process creates an enormous amount of data. Indeed, advances in computational imaging that allowed data from different Event Horizon telescopes to be combined into a single image was how scientists managed to create a photo of a black hole at the heart of distant galaxy Messier 87 in April this year.
Interestingly, the technology of computational photography is increasingly used in newer handsets, such as the Apple iPhone 11, where a single photo is created from three camera modules working simultaneously. Armed with three high-definition cameras, these new generation handsets are able to capture better-quality images even in low-light conditions.
On the whole, the role of cameras in the digital era is currently being redefined. Beyond producing videos and still images for personal and social media content, cameras are now use extensively in the areas of IoT, machine-to-machine communication, (such as in autonomous driving), AR/VR, security features recognition and others (Chart 6).
Chart 6: Role of cameras in the data era redefined
Source: ASM Pacific
All in all, datasets globally are expanding in a rapid way and this creates ample long-term opportunities for investors in the technology sector. As one of the world’s most valuable resources in today’s digital age, data is indeed the “new oil” that is powering the growth of technology companies.
Asia, and notably China, stands to benefit from the unfolding data revolution. For example, China’s massive e-commerce market, which dwarfs its global peers, will generate extensive demand for data as the country’s data centre capacities are upgraded to hyperscale levels. The reconfiguration of Asia’s data landscape may bode well for sectors pertaining to e-commerce, industrial IoT, gaming and tech hardware.
1 An exabyte is 1 billion gigabytes.
2 A zettabyte is about 1,000 exabytes or one trillion gigabytes, and a yottabyte is approximately 1,000 zettabytes. *IDC estimates
3 Source: Cisco’s Visual Networking Index; Forecast and Trends, 2017–2022 White Paper, February 2019