The tech industry has been shaped by four significant waves of development beginning with the computer, moving on to personal computing, the internet and later, the mobile. Data-driven computing, the fifth wave, will herald an era of one trillion connected devices.
It’s an exciting time to be in the tech industry. So far, the industry has been characterized by four waves of significant development and the computing performance of devices has increased rapidly due to the technological advancement coming from these developments.
The data-driven computing era which constitutes the fifth wave will bring a paradigm shift in the way we live and perceive things. Technologies like the Internet of Things (IoT), 5G and artificial intelligence (AI) will bring in the required change.
“The fifth wave will have a transformative impact like none other. The spotlight is on connectivity. 5G is a world of one trillion connected devices. ARM has delivered 100 billion chips in 26 years, but the next 100 billion chips will be shipped in four years. This is because the power of computing has grown exponentially and the demand from connected devices will scale rapidly,” said Chet Babla, IoT Services Group, Arm, speaking at the ARM Tech Symposia India 2018.
A lot of growth will be driven by Asian economies. Within Asia, India will be a prominent growth driver of 5G. It’s largely because the Government of India (GoI) has envisioned 100 Smart Cities. AI has a role to play in the layout of streets and devices used for regular day-to-day operations. Machine learning (ML) will be leveraged to programme devices and services in Smart Cities. India will be a key contributor, as AI and ML will be used to solve everyday real-life problems.
“Asia-Pacific will lead the pack in the 5G segment by 2025, according to the latest edition of GSMA’s Mobile Economy Report. Billions of mobile devices will be connected,” Babla added.
Forthcoming estimates indicate that 40m new servers will be deployed for video surveillance. This new dimension will be facilitated by data, which will flow from the cloud to the endpoint and the movement will be upstream.
Besides 5G, another upcoming trend is that the mobile is being increasingly designed to give a laptop performance. The computing performance of the mobile and laptop browser is increasing. There’s a bigger thrust on tenets like reliability, latency, privacy, and security. The buzzword is edge performance, facilitated by AI and ML.
High-definition video sensors used in mobiles require multiple smart IoT applications. Arm unveiled its mobile ML processor platform earlier in the year, described as Project Trillion. The company feels ML needs to be incorporated into mobiles for rendering top-class computing solutions.
“We believe that mobiles should offer a laptop type of productivity. Year-on-year, our performance is increasing and is expected to reach 15 percent by 2020. Put all this together and our performance will increase by 2.5 times,” explained Babla.
IoT is another dimension of the fifth wave. Making sense of the volume of data and velocity at which it is generated is a challenge. The general consensus is to reduce the complexity of IoT.
“By 2035, it is expected that IoT data will connect one trillion devices. The value of IoT can be realized only if we have usable data. IoT has its share of technical challenges that include harnessing data, managing data, connecting devices and developing devices,” reasoned Chris Porthouse, IoT Services Group, Arm. It’s important to reduce friction in each of these stages for better device management and connectivity.
Data for devices is valuable only if it is secure. Security happens only if every segment of the ecosystem is strengthened, right from device to cloud to data analytics. A reality check reveals that data supply the world over is still fragmented. Streaming data requires real-time learning and intelligence-based research. A large part of the data is unstructured, scattered and siloed. Hence it requires intelligent tools. High-performance computing can happen provided bottlenecks in IO (input/output) are removed.
A convergence of technologies along with IoT device-data investments will help the industry to scale up to the demands of the upcoming data-driven computing era.