The computing power industry is thriving

The rapid rise of emerging fields such as scientific research, artificial intelligence, digital twins, and the metaverse is driving the rapid expansion of global computing power. This growth is accelerating multi-faceted innovation in computing technologies and products, reshaping and restructuring the industry landscape. As a result, computing power has become a new engine for global digital economic development and a strategic focal point in national-level competition.

Steady growth in computing power scale

The global computing power scale continues to grow steadily at a high level. In the era of the digital economy—defined by ubiquitous perception, connectivity, and intelligence—the total volume of global data and computing power is maintaining a trend of sustained high growth. According to data research, by the end of 2023, the total global computing power reached 910 EFLOPS (FP32), representing a year-on-year increase of 40%, while intelligent computing power grew by 136% year-on-year.

It is projected that by 2030, humanity will enter the Yottabyte (YB) data era, with global computing power reaching 56 ZFLOPS and an average annual growth rate of 65%. Among this:

• Basic computing power will reach 3.3 ZFLOPS with an average annual growth rate of 27%;

• Intelligent computing power will reach 52.5 ZFLOPS with an average annual growth rate of over 80%;

• Supercomputing power will reach 0.2 ZFLOPS with an average annual growth rate exceeding 34%.

Diverse demands are accelerating the upgrade and diversification of computing power. The rapid rise of emerging fields such as artificial intelligence, scientific research, and the metaverse has raised new demands on computing capabilities. For example, in the case of the metaverse, Intel predicts that computing power will need to increase by a factor of 1,000, while NVIDIA estimates that immersive real-time rendering will require computing power a million times greater than currently available.

In terms of basic computing power, IDC’s latest data shows that the global Infrastructure as a Service (IaaS) market is expected to reach USD 262.68 billion in 2024, up approximately 12.3% from USD 233.91 billion in 2023. Meanwhile, the Platform as a Service (PaaS) market is also showing strong growth momentum. According to IDC, the global PaaS market is expected to reach USD 159.6 billion in 2024, a 37.08% increase from USD 116.4 billion in 2023. Combined, the global IaaS and PaaS markets are projected to reach a total of approximately USD 422.28 billion in 2024, up around 20.5% from USD 350.31 billion in 2023.

Cloud computing is poised to become the mainstream general-purpose computing model of the future, providing foundational support for big data, artificial intelligence, 5G, and other emerging technologies, as well as serving as a driving force for digital transformation and intelligent upgrades across industries.

In the realm of intelligent computing, there is a growing contradiction between the demand for processing massive and complex datasets and the limited supply of single-source computing power. With the rapid increase in global data volume, over 80% of data is unstructured (e.g., text, images, voice, video). As Moore’s Law and Dennard scaling slow down, the annual performance improvements of chips—such as CPUs—have dropped to under 15%, making it increasingly difficult to meet the computing needs of unstructured data like video and images. This has created an urgent demand for diversified intelligent computing power.

The Frontier supercomputer, developed by the U.S. Oak Ridge National Laboratory (ORNL), achieved a performance of 1.102 EFLOPS in the Linpack benchmark test, surpassing Japan’s Fugaku to become the world’s first officially confirmed exascale supercomputer. This milestone marks the formal entry of the global supercomputing industry into the exascale era.