Chinanews, Beijing, December 11 (Reporter Zhang Su) Intelligent computing is the underlying source of driving the development of artificial intelligence.
Driven by huge demand, the construction of smart computing infrastructure is regarded as a hot spot in the "new infrastructure".
However, how to "build, manage, and use it well" has attracted attention from all parties.
"The smart computing center is not only the foundation of the digital economy, but also an important practitioner and promoter of the green economy. We hope to drive the open and coordinated development of the industrial ecology through the '5A' smart computing infrastructure." December 10, Sugon Advanced Vice President Ren Jingyang said at an event in Beijing.
According to statistics, in the first half of this year, there were more than 300,000 new artificial intelligence companies in China, a year-on-year increase of 88%.
According to the prediction of Roland Berger International Management Consulting Company, from 2018 to 2030, the demand for artificial intelligence computing power will increase by 390 times.
The "Research Report on the Development of New Generation Artificial Intelligence Computing Infrastructure Development" released by the National Industrial Information Security Development Research Center believes that artificial intelligence has become the focus of competition among countries in the world. The intelligent computing center is an important underlying infrastructure form for the development of the artificial intelligence industry. It has also been pushed to the heights of development.
Ren Jingyang said that the "new infrastructure" represented by intelligent computing infrastructure has become an important starting point for the development of the digital economy.
But because its construction is still in its infancy, it naturally faces many challenges.
For example, as a public computing platform infrastructure, what computing power should the smart computing center provide?
How does the smart computing center support the "dual carbon" initiative to reduce energy consumption?
How can we cooperate with the upstream and downstream forces of the industry to build a smart computing center?
How can the smart computing center support the local economy and society?
...
In July this year, the Chinese Academy of Sciences Artificial Intelligence Industry-University-Research Innovation Alliance released a new generation of artificial intelligence computing platform, which provides a reference basis and construction standards for the construction of intelligent computing centers from the dimensions of infrastructure, construction path, and price model.
Sugon is a major participant in the new generation of artificial intelligence computing platform, and in September it proposed the core capabilities of "5A" intelligent computing.
According to Ren Jingyang, "5A level" refers to the five dimensions of openness, integration, greenness, inclusiveness, and service.
He took "openness" placed in the first place as an example, as a public computing power service platform, compatible with mature mainstream hardware and software technology ecology, it is a basic and necessary capability.
The "5A-level" intelligent computing center aims at diversified collaboration. Through the comprehensive openness and compatibility of chips, algorithms, frameworks, and models, it builds a diversified and integrated infrastructure to achieve the maximum ease of use of the computing power base and reduce migration costs.
As a smart computing infrastructure with public attributes, it is not uncommon for the construction investment to be too high.
The “5A-level” intelligent computing infrastructure has continuously optimized the cost of computing power through a variety of strategies to achieve a comprehensive construction cost that is 30% lower than the existing market price.
The "Research Report on the Development of a New Generation of Artificial Intelligence Computing Power Infrastructure" recommends that in the process of building a smart computing center, the competent authority should focus on strengthening top-level system planning and standard construction, and local governments should improve their understanding and application of computing power economy. Enterprises should actively explore the research and development of key artificial intelligence technologies, the construction of ecological applications, and the innovation of service models.
"Dawning's intelligent computing infrastructure has been completed and planned in many cities across the country." Ren Jingyang said, hoping to build the era of artificial intelligence with scientific formulas to promote the ecological prosperity of the regional artificial intelligence industry and the high-quality development of the digital economy.
(Finish)