Promoting the High-Quality Development of the Green Computing Power Industry
China should promote the high-quality development of the computing power industry based on green and low-carbon principles by optimizing its spatial layout, building a complete industrial chain, and deepening coordination between computing power and electricity. Demand for artificial intelligence training can be transferred in an orderly manner to western regions rich in green electricity. Meanwhile, coordinated efforts should be strengthened to achieve breakthroughs in chips, large-model infrastructure, and algorithm frameworks, and to build zero-carbon intelligent computing centers powered entirely by green electricity. Standards for energy-efficiency assessment and green electricity traceability also need to be improved to enable real-time matching and flexible adjustment between computing workloads and electricity supply.
Body · AI translation
During an inspection tour in Ulanqab, Inner Mongolia Autonomous Region, General Secretary Xi Jinping stressed: “The computing power network is one of the ‘six networks’ planned and being built by the state. The computing power industry is currently in a stage of accelerated development and has broad prospects. I hope everyone will earnestly implement the requirements for efficiency, sustainability, and security, with an emphasis on coordinating computing power with electricity and water resources, properly developing and operating computing power center projects, actively advancing the green computing power industry, and making greater contributions to building a Digital China and fostering new quality productive forces.” The State Council’s Action Plan for Carbon Peaking During the 15th Five-Year Plan Period explicitly calls for “advancing the green and low-carbon transformation of computing power facilities,” requiring better-coordinated planning of computing power facilities and renewable energy and faster development of green computing power facilities.
Developing the green computing power industry is essential to building a green and intelligent digital ecological civilization. From the perspective of global development trends, developing this industry is a strategic choice for securing the initiative in future digital competition and overcoming the constraints imposed by computing power’s energy consumption. As artificial intelligence technologies rapidly evolve worldwide and computing capacity expands quickly, pressure from energy consumption and carbon emissions is rising in tandem. Global competition in computing power is no longer confined to its scale; the overall capacity to supply green computing power is increasingly becoming an important competitive dimension. From the perspective of China’s overall development, developing the green computing power industry is intrinsically necessary for putting the philosophy of green development into practice and cultivating and expanding new quality productive forces. The industry can provide clean, low-carbon computing power for intelligent manufacturing, smart governance, and digital industries, laying a solid foundation for new forms of the intelligent economy. We must seize the opportunities, incorporate the green and low-carbon philosophy throughout the development of the computing power industry, promote the high-quality development of the green computing power industry, and build a green and intelligent digital ecological civilization.
Optimize the spatial distribution of green computing power. We should accelerate the formation of a green computing power supply structure aligned with the East Data, West Computing project, guide “computing power to follow electricity,” and systematically shift the high-load, energy-intensive demand for AI training to western regions rich in green electricity. Adopting a coordinated nationwide approach, we should balance the use of western electricity in the west with its transmission to the east, promote coordination among energy development, industrial relocation, and clean energy deployment, and foster the integrated development of advanced manufacturing, computing power, hydrogen energy, and other industries. We should establish interregional computing power supply corridors and explore the creation of interregional coordination mechanisms covering the entire process of computing power scheduling, network transmission, computing-electricity integration, operational services, transaction settlement, and revenue distribution. Beijing, for example, has issued an implementation plan for computing infrastructure development, continued to advance such development, and strengthened the mechanism for the integrated and coordinated development of computing power across Beijing, Tianjin, Hebei, and Inner Mongolia. Greater support should also be provided by the core digital economy regions in the east for the construction of large-scale data centers in the west, further deepening east-west coordination in computing power.
Build a complete green computing power industrial chain. We should promote technological innovation in green computing power, deepen collaboration among industry, academia, and research institutions, and conduct coordinated research on key areas such as AI chips, foundational architectures for large models, and algorithm frameworks. We should expand the use of green, energy-saving measures such as photovoltaic power generation and waste-heat recovery, improve the energy and carbon efficiency of computing infrastructure, and create benchmark, demonstrative zero-carbon intelligent computing centers powered entirely by green electricity. Attention should be paid to coordinating computing power with water resources and reducing data centers’ water consumption through their technical architecture. We should accelerate energy-saving and carbon-reduction upgrades at inefficient data centers and consolidate and renovate old, small, and scattered data centers. We should vigorously develop midstream and downstream computing power industries, including cloud computing, cloud storage, the industrial internet, and smart terminals, and establish a number of green computing power industrial parks. We should diversify and innovate green computing power application scenarios, promote its application in industry, culture and tourism, transportation, healthcare, education, urban governance, and other fields, and develop a number of highly demonstrative, broadly influential application scenarios empowered by intelligent computing, thereby accelerating the formation of a complete green computing power industrial chain.
Deepen the coordinated development of computing power and electricity. Supported by digital technologies, intelligent algorithms, and information networks, such coordination is a new form of infrastructure that enables computing facilities and power systems to operate together efficiently while facilitating real-time alignment and optimized allocation between computing loads and electricity supply. It is a key means of cultivating the overall capacity to supply green computing power and must be advanced in multiple areas, including planning, standards, scheduling, and markets. We should scientifically plan the integrated deployment of “green electricity and computing power,” strengthen cross-industry integration between green electricity and computing power, continuously expand coordinated application scenarios, and enable their coordination to better empower industrial development. We should improve the operating mechanisms for computing-electricity coordination, refine standards and specifications for areas such as the energy-efficiency assessment of computing facilities and the traceability of green electricity, fully tap the potential for adjusting computing loads across time and space, remove data barriers between computing power scheduling and electricity dispatch, and promote real-time matching and flexible adjustment between computing loads and electricity supply, enabling more entities to use green computing power at lower cost and with less difficulty.
(The author is a specially appointed research fellow at the Fujian Research Center for Xi Jinping Thought on Socialism with Chinese Characteristics for a New Era.)
AI Trends