Advertisement
Advertisement
China’s Communist Party
Get more with myNEWS
A personalised news feed of stories that matter to you
Learn more
China’s national data network aims to address digital imbalances between the more prosperous areas of eastern China and the energy-rich west. Shutterstock

China’s new data agency plans fast track to powerful national computing network

  • China – second only to the US in terms of aggregated processing power – sets 2025 deadline to scale up capacity by half
  • At first work conference since its formation last year, NDA vows to spend government funds well as it connects regional computing centres
China will accelerate construction of a nationwide integrated computing power network, its top data regulator vowed at its first annual meeting as Beijing seeks to boost innovation in a technology competition with the United States.
At the two-day national data work conference which ended on Tuesday, National Data Administration (NDA) head Liu Liehong vowed to “optimise data infrastructure distribution” and “accelerate construction of the computing power network”.

“We are working to improve the distribution of general purpose, intelligent and supercomputing power, increase the synergy of east-middle-west regions, support the integration of computing power and green electricity, as well as balance between its development and security,” he said, according to an NDA post on social media platform WeChat.

The conference took place just months after the NDA was inaugurated at the end of October, in response to President Xi Jinping’s calls to drive digital development and innovation in a competition with the US for hi-tech dominance.

The nationwide computing power network, which will be up and running by next year, is designed to unite computing centres across the country, addressing regional digital imbalances between the more prosperous areas of eastern China and the energy-rich west, the NDA has said.

China is spending billions on a national computing network. Its data chief says why

China, which is second only to the US in aggregated computing power, aims to scale up its capacity by half by 2025. Demand for computing power, particularly the GPU power necessary for artificial intelligence learning, has been surging in recent years.

The graphics processing unit (GPU) was originally designed to accelerate computer graphics and image processing in games. But it can also perform high-speed calculations and has become integral to the development of large language models.

But the US Commerce Department tightened hi-tech export controls in October, citing national security concerns. The sanctions prevent Nvidia – which has a near monopoly on GPUs – from exporting its advanced chips to China.

Xue Lan, professor and director of Tsinghua University’s Institute for AI International Governance, said in February that many Chinese companies had been affected by the Nvidia ban.

“GPU power has become a bottleneck [for China’s AI ambitions],” he told a forum in Hangzhou, provincial capital of Zhejiang, in eastern China.

Xue noted that another problem faced by China’s technology companies and researchers is “the relatively low quality” of data. “China has a large amount of data, but it has not been industrialised. Standardised data service providers are few,” he said.

05:03

How does China’s AI stack up against ChatGPT?

How does China’s AI stack up against ChatGPT?

According to the NDA’s WeChat post, Liu also vowed at this week’s conference to make good use of government investments and facilitate the production of data and its circulation.

The agency had heard Xi’s “concerns on data development and security” and put the marketisation of data resources front and centre in its efforts to support China’s “high quality” development, he said.

This year, the NDA will prioritise improving data ownership, circulation and transaction policies and increasing data development and utilisation, it said.

The NDA will also maintain “high alert” on data security and promote international cooperation by optimising regulations to facilitate cross-border data flows, according to its social media post.

1