In the era of intelligent computing, a Chinese answer

This year, as AI big models have been rapidly recognized, the global intelligent industry has ushered in a new peak of development. Major technology companies around the world have deployed big models, and the domestic "hundred-model war" has even seen a grand event.

With the rapid rise of large models, a problem that has plagued the AI ​​industry for many years has been further exposed: the supply of specialized computing power for AI tasks remains in short supply as demand continues to grow.

For a while, countries around the world generally experienced a "computing power panic" and "hard to get a card" phenomenon, and there was even a strange phenomenon that European and American technology companies began to use GPU resources for financing.

These phenomena clearly tell us that in the intelligent era, computing power is productivity and capital. Facing the intelligent era, how can we solve the challenges of computing power shortage and computing power shortage? How can we develop intelligent computing power technology in a long-term and healthy way? How can we make intelligent computing power a driving force for technological self-reliance?

These questions require a definite answer.

On August 18, the 2023 China Computing Power Conference, China's first provincial-ministerial joint conference with the theme of empowering data center computing power, opened in Yinchuan, Ningxia. With the theme of "Leading new industrial trends with computing power, empowering high-quality development", this conference detailed the development prospects and latest achievements of China's computing power industry.

Facing the surging intelligent era, we can find and confirm the "Chinese answer" to intelligent computing power at the China Computing Power Conference.

Definition and Trend of Intelligent Computing Power

Faced with the surging wave of AI, perhaps everyone has heard of the rising popularity of GPUs, but perhaps few people can explain in detail what exactly is intelligent computing power?

From a basic technical perspective, intelligent computing power, also known as artificial intelligence computing power, is the computer system capability that provides artificial intelligence algorithm model training and model operation services for artificial intelligence applications.

As the importance of AI technology continues to increase, the value of intelligent computing power as infrastructure is also increasing. It can be said that intelligent computing power is the foundation of intelligent applications and intelligent industries, and is a prerequisite for developing an intelligent economy and building an intelligent society.

Regardless of the aspects, intelligent computing power is in a period of rapid explosion. By the end of 2022, the total global computing power will reach 650 EFLOPS, of which the intelligent computing power will be 142 EFLOPS, accounting for an astonishing 21.9%, an increase of 25.7% compared with last year. It can be said that in the entire computing industry landscape, intelligent computing power is currently the fastest growing, most in demand, and most concerned.

Specifically speaking of the development of China's intelligent computing power, we can see that by the end of 2022, China's total computing power will be 180 EFLOPS, and the scale of intelligent computing power will be 41 EFLOPS. In other words, the scale of intelligent computing power accounts for 22.7%, an increase of 41.4% compared with last year. This growth rate far exceeds the average growth of global intelligent computing power. In other words, China's intelligent computing power industry is currently in a development stage with high demand, fast growth and high value. It can be seen that semiconductor and computing companies have targeted the wave of intelligent computing power and launched targeted products and solutions. For example, H3C has launched AI servers specifically for large model training and 51.2T, 800G CPO silicon photonic data center switches, as well as Aofei computing power platforms that support large computing power scheduling.

After understanding the definition and development speed of intelligent computing power, the next question is, what application value will intelligent computing power ultimately bring?

From the current situation, the hottest technology trends such as Metaverse, autonomous driving, AIGC, and digital twins all seem to be inseparable from the support of intelligent computing power. For example, AIGC is currently a very popular field, which has set off a wave of investment and construction around the world. Behind AIGC is the organic integration of core elements such as computing power, data, and algorithms. The larger the model used, the higher the computing power required. Judging from the current trend, AIGC's demand for intelligent computing power will be stronger in the future, thereby reshaping the value of industries related to intelligent computing power.

At the 2023 China Computing Conference, we can see that all walks of life are paying attention to the solutions and supply of intelligent computing power, and are looking forward to obtaining stronger support from intelligent computing power. This is because intelligent computing power is the source of intelligent technology and applications, and the foundation of intelligent innovation.

Next, we can go deeper and explore more possibilities in the development of intelligent computing power.

Challenges and opportunities facing intelligent computing power

As far as any technology industry is concerned, it must continue to develop in the cycle of "encountering challenges-solving problems-industry upgrading", and intelligent computing power is no exception. It can be seen that with the huge demand for intelligent computing power emerging in the global technology industry, there has been a significant shortage of supply in this field, which has led to phenomena such as "computing power panic" and "computing power difficulties". At the same time, with the continuous development of AI technology, problems such as energy consumption, security, and ecological cooperation have gradually been exposed. However, only with challenges can there be development goals and opportunity nodes for related companies and practitioners.

The 2023 Intelligent Computing Power Development White Paper points out that intelligent computing power development faces five major challenges. At the same time, the White Paper also provides detailed solutions to each challenge to help the industry consolidate consensus and capture development windows.

For example, computing power demand is the core and most significant challenge of intelligent computing power. The sharp increase in demand for intelligent computing power is a common problem faced by the global AI field. In order to cope with the shortage of computing power, including intelligent computing power, and promote the intensive development of social computing power, China has creatively proposed and implemented the East-West Computing Project. At the same time, it should be noted that the linkage challenges between data centers have resulted in the efficiency of computing power resource allocation still needing to be improved. Solving these problems constitutes a major opportunity in the development of intelligent computing power and the continued advancement of East-West Computing.

To this end, the White Paper proposes a solution to upgrade the supply of computing resources by building intelligent computing centers and cloud computing centers, and to dispatch idle computing power around data sources through the network. Combined with solutions such as improving the efficiency of computing power use, developing new network architectures, and establishing a unified computing power dispatching platform, the value of East-West computing can be maximized to meet the huge demand for intelligent computing power in all sectors of China's industry.

In addition to the demand for computing power, the White Paper also proposes a series of challenges, including the increasing energy consumption challenges; the continuous improvement of algorithm complexity; the data privacy and security issues faced by AI models and AI computing; and the difficulties in the cooperation of the industry ecosystem related to intelligent computing power. It also provides specific solutions and suggestions for these challenges. It can be said that at this stage, intelligent computing power is still in the initial stage and golden period of rapid development. Only by identifying the challenges and exerting efforts efficiently can we maximize the dividends in the development of intelligent computing power.

China’s answer to the future

After summarizing the definition and examining the challenges, the next most critical question is: How should China's technology industry break through the development of intelligent computing power? How to give an answer that conforms to the trend of the times.

To this end, the White Paper provides an in-depth insight into the future development of intelligent computing power and provides development suggestions for various related industrial fields. It points out that there will be five basic trends in the future development of intelligent computing power:

1. Artificial intelligence is penetrating at an accelerated pace, and diversified scenarios are giving rise to diversified computing power demands.

In the future, AI technology will not exist alone, but will deeply penetrate into scenarios and applications. Therefore, heterogeneous and diversified computing needs will continue to increase, and cloud-edge-end collaborative intelligent computing will become the mainstream. Driven by these factors, we will enter a new era of diverse computing.

2. Driven by policies, low-carbon development of intelligent computing power has become a rigid requirement.

The development of intelligent computing power happens to be in sync with the implementation of the dual carbon goals. Therefore, in the future, under the guidance of policies, a more blanket approach to computing power acquisition will be adopted to meet AI needs. Low-carbon technology will also become a core competitiveness of the computing and AI industries.

3. Edge intelligence came into being, and edge computing and artificial intelligence were integrated and developed.

Relatively speaking, obtaining intelligent computing power from the cloud will generate great dependence on the network environment, while localized intelligent computing power faces high hardware costs. Therefore, in the future, it is very likely that the way to obtain intelligent computing power will be migrated to the edge side, so AI technology will develop synchronously with the edge computing industry.

4. The construction of intelligent computing centers is accelerated to meet the demand for high-quality computing power.

In the tide of East-West computing and the new generation of data centers, intelligent computing centers that provide heterogeneous computing resources will become a new outlet. Intelligent computing centers will provide exponential improvements in intelligent computing power, achieving performance upgrades and energy consumption reductions in AI scenarios.

5. The scale of the model continues to expand, and massive and diversified data urgently requires massive computing power.

Since the rise of deep learning, a clear rule has been that the scale of models is getting bigger and bigger, and the intelligent emergence effect brought by the scale of models is becoming clearer and clearer. There is reason to believe that huge models with super-large parameter numbers will appear next. The current large models may just be the beginning, and intelligent computing power must be prepared for this.

Driven by these five trends, the intelligent computing industry will usher in a certain future development. First, the demand for computing power will continue to explode, and the momentum of industrial development and supply chain challenges will continue to increase; second, the localization process of intelligent computing power will inevitably accelerate, and full-stack autonomous intelligent computing power will become the focus and strategic pillar of future industrial development; in addition, intelligent computing power will be integrated into thousands of industries to create value, which will provide impetus for the development of intelligent computing power solutions and industrial ecology in various fields.

To this end, the White Paper puts forward suggestions for the development of intelligent computing power in terms of industry, technology, and standards.

For example, in terms of industry, it is recommended to formulate a plan from the national strategic level and take multiple measures to promote the healthy and orderly development of intelligent computing power. These include building a unified intelligent computing service center and incubation platform, strengthening financial support for related industries, and promoting the green development of intelligent computing power.

In terms of technology, we should increase the research and development of intelligent computing power, especially the innovation of key technologies. In terms of standards, we should accelerate the construction of open standards and transform diversified computing power into dispatchable computing power resources.

According to IDC's forecast, China's intelligent computing power will continue to grow rapidly, and is expected to reach 1271.4 EFLOPS by 2026, with a compound growth rate of 52.3% in the next five years. It can be said that intelligent computing power is ushering in its golden development period in China, and all related technical fields and industrial ecological roles will usher in historic development opportunities.

The great river of intelligent computing power requires us to understand, face and explore. In the end, we will sail on it and give China's answer to intelligent computing power and the intelligent era.