Welcome to the channel of the little demon fairy, please brothers and sisters to move your rich little hands to follow, like, comment and forward, thank you! Here, I wish you all family harmony, filial piety, smiles, and happiness in your heart!
Zhang Ping'an, CEO of HUAWEI CLOUD: 7nm is not a panacea, and new concepts are needed for AI development! 》
In today's rapid development of science and technology, chip technology is often regarded as an important indicator to measure a country's scientific and technological strength. However, at a time when the global chip manufacturing process nodes are shrinking and 3nm and 5nm have become hot topics, Chinese technology giant Huawei has thrown out a thought-provoking opinion: we must abandon the concept that "we cannot develop without the most advanced chips". This statement, like a thunderbolt, has triggered extensive discussion and reflection in the scientific and technological circles.
In July 2024, Zhang Ping'an, Executive Director of Huawei and CEO of HUAWEI CLOUD, made a remark that brought new ideas for the development of science and technology in China and around the world. He bluntly pointed out: "China's AI development is inseparable from the innovation of computing infrastructure, and we must abandon the concept of 'no development without the most advanced chips'." These words are not only a deep insight into the current status of scientific and technological development, but also a bold prediction of the future development direction.
In today's world, the advancement of chip technology seems to have become a race with no end in sight. From the original micro-scale process to today's nano-scale process, the continuous advancement of the chip manufacturing process has brought us more powerful computing power and more energy-efficient electronic devices. However, as process nodes continue to shrink, the difficulty and cost of chip manufacturing are also rising dramatically. Against this backdrop, Huawei's new ideas have undoubtedly opened a new window of thought for us.
Zhang Ping'an said frankly: "No one will deny that we are facing the problem of limited computing power in China. This sentence illustrates the challenges that China is currently facing in the field of high-end chips. Subject to various factors, China does have a certain gap in the most advanced chip manufacturing process. However, Zhang's next words give hope: "But we can't just rely on having AI chips with advanced manufacturing process nodes as the ultimate foundation of AI infrastructure." "
This remark actually reveals an important fact: the more advanced the chip manufacturing process, the better, the key is how to rationally use the existing technology to maximize the efficiency. Zhang Ping'an pointed out that Huawei's innovation direction is to release the AI computing power demand on the device side to the cloud through optical fiber and wireless networks, and obtain seamless AI computing power through device-cloud collaboration. This innovative approach not only allows end-side devices to maintain rich functions, but also greatly reduces power consumption and dependence on chips.
Huawei's approach reflects an important concept of technological development: technological innovation should not be confined to a single field, but should be a systematic process. In the case of limited chip technology, we can improve the overall performance by optimizing the system architecture, increasing network bandwidth, and improving energy efficiency. This kind of comprehensive innovation may bring new opportunities for the development of science and technology in China and even the world.
However, we also need to be soberly aware that this does not mean that we can give up the pursuit of advanced chip technology. On the contrary, we should strive to catch up with advanced chip technology while also actively exploring other innovation paths. Only in this way will we be able to take an advantageous position in the fierce global competition in technology.
Huawei's thinking has brought us some profound enlightenment:
First, we need to re-examine the nature of technological development. The purpose of scientific and technological development should not only be to pursue the improvement of a specific index, but to solve practical problems and improve the quality of human life. In this process, we need to consider all factors holistically, rather than blindly chasing a single metric.
For example, in the field of artificial intelligence, while more advanced chips can provide more computing power, this does not mean that we have to rely on the most advanced chips. By optimizing algorithms, improving model structures, and leveraging distributed computing, we can also achieve powerful AI capabilities. In fact, the core value of many breakthrough AI applications often lies in innovative algorithms and models, rather than relying solely on the computing power of hardware.
Second, we need to recognize the diversity of innovation. Technological innovation should not be confined to a specific field, but should be an integrated process. In the case of a limitation in one area, we can make up for it with innovation in other areas, and maybe even find a better solution.
For example, in the field of the Internet of Things, we can improve the overall performance of the system by optimizing the communication protocol between devices and improving the efficiency of data processing, rather than relying solely on the computing power of each device. In this way, we can build robust IoT systems with a large number of low-cost, low-power devices, rather than relying on a small number of high-performance devices.
Third, we need to be open and inclusive. In today's globalized world, the development of science and technology is no longer a matter for a country or a company to do alone. We need to strengthen international cooperation and jointly promote scientific and technological progress. Even in the face of challenges and limitations, we should keep an open mind and seek opportunities for collaboration.
For example, in the field of chip design, we can use open-source hardware to gather global wisdom to jointly promote technological progress. A good example of this is the RISC-V instruction set architecture, which is an open source approach that enables developers around the world to participate in chip design, thereby advancing the entire industry.
Finally, we need to develop a long-term strategic vision. The development of science and technology is a long-term process that cannot be achieved overnight. We need to develop a long-term development strategy that looks at both the current issues and the way forward. Only in this way can we go further on the road of scientific and technological development.
For example, in the field of clean energy, although current technologies may not be able to completely replace traditional energy sources, we still need to continue to invest in research and development to prepare for the future energy transition. Similarly, in the field of quantum computing, while the current technology is far from mature, we still need to stay invested and prepare for the technological revolution that may arise in the future.
Huawei's innovative thinking actually points us to a new development path. In the case of limited chip technology, we can improve the overall performance by optimizing the system architecture, improving network performance, improving energy efficiency, and other aspects. This kind of comprehensive innovation may bring new opportunities for the development of science and technology in China and even the world.
For example, in the field of cloud computing, we can improve the performance of cloud services by optimizing the design of data centers and improving network transmission efficiency. In this way, we can provide powerful and feature-rich cloud services with relatively outdated hardware. In fact, many cloud service providers have already made successful attempts to deliver high-performance cloud services by optimizing software architectures and network structures, rather than relying solely on state-of-the-art hardware.
In the field of edge computing, we can reduce the pressure on the central server by optimizing the distribution of computing tasks and placing a part of the computing tasks on the edge device. This not only improves the responsiveness of the entire system, but also reduces the need for network bandwidth, resulting in more efficient computing.
In the field of artificial intelligence, we can transform large and complex AI models into small and efficient models through technologies such as model compression and knowledge distillation, so that powerful AI functions can be realized on ordinary hardware. This approach not only reduces the dependence on high-end hardware, but also allows AI technology to be more widely applied to various end devices.
This innovative idea can not only help us overcome the limitations of chip technology in the short term, but also may bring more long-term benefits. Through systematic innovation, we may find solutions that are more efficient, more environmentally friendly, and more relevant to real needs. This will not only benefit China's scientific and technological development, but also contribute Chinese wisdom to global scientific and technological development.
However, we also need to recognize that this innovative approach is not without its challenges. First, it requires a more holistic and systematic mindset that optimizes system performance from a holistic perspective, rather than simply relying on the performance of individual components. This places higher demands on our engineers and researchers.
Second, this innovative approach may face some short-term difficulties. For example, in certain application cases, the most advanced chips may indeed deliver significant performance gains, while other optimizations may not fully compensate for this gap. Therefore, we need to find a balance between short-term interests and long-term development.
Thirdly, this innovative idea may face some conceptual resistance. For a long time, we have been accustomed to measuring the level of technology with the most advanced technical indicators, and it will take time to change this mindset. We need to change this mindset through ongoing practice and success stories.
Still, there are reasons to be optimistic about this innovative approach. Historically, many major technological breakthroughs have been made under constrained conditions. It is these limitations that force us to think outside the box and explore new possibilities. Huawei's approach may be an opportunity to rethink the nature and direction of technological development.
Overall, this new perspective put forward by Huawei has opened a new window of thought for us. It reminds us that the development of science and technology should not simply be equated with the improvement of a specific indicator, but should be a comprehensive and systematic process. In this process, we need to give full play to the spirit of innovation and find the most suitable development path for ourselves.
As Zhang Ping'an said: "We must abandon the notion that 'we cannot develop without the most advanced chips'." "This is not only a rational understanding of the current situation, but also a positive outlook for future development. On the road of scientific and technological development, we need to maintain an open and inclusive attitude, be brave in innovation and dare to make breakthroughs. Only in this way can we continue to advance in the fierce global scientific and technological competition and make greater contributions to the progress of human society.
Huawei's approach has also brought inspiration to other enterprises and research institutions. In the face of technical limitations, we should not passively wait, but actively look for breakthroughs. Through systematic innovation, we may find more opportunities for development.
In the future development of science and technology, we need to maintain this open, inclusive and innovative attitude. We should not only strive to catch up with the most advanced technology, but also be good at using existing technical resources to solve practical problems through innovative thinking and systematic methods. Only in this way can we go further on the road of scientific and technological development and create greater value for human society.
Finally, we need to realize that the ultimate purpose of scientific and technological development is to solve the practical problems faced by mankind and improve the quality of human life. Whether it's state-of-the-art chips or innovative system architectures, they're just the means to make this happen. We need to always maintain this goal-orientedness, and always strive for the well-being of mankind while pursuing technological progress. This is the true meaning of scientific and technological development.
