清华大学在芯片研究领域取得重大突破，为高性能计算带来全新可能。据11日的最新消息，该校电子工程系副教授方璐与自动化系戴琼海院士的课题组，针对大规模光电智能计算的挑战，成功研发出分布式广度光计算架构，命名为“太极”（Taichi）的干涉-衍射异构集成芯片。
“太极”芯片摒弃了传统的电子深度计算模式，开创性地采用光子计算，实现了160 TOPS/W的通用智能计算能效，这一成果在大模型通用人工智能的背景下显得尤为突出。在当前算力需求日益增长的时代，这一科研突破为高性能算力的提升提供了创新灵感、架构和路径。
这一重要进展已发表在国际权威期刊《科学》的最新一期上，标志着中国在光子计算领域的领先地位，以及在全球高科技竞赛中迈出的坚实一步。清华大学的这一创新成果不仅将推动科研领域的进步，也将对未来的云计算、人工智能和大数据处理等领域产生深远影响。
英语如下：
**News Title:** “Tsinghua University Breaks Through in Chip Technology with Photonic Computing Chip ‘Taichi’ Featured in ‘Science’ Magazine, Empowering High-Performance AI”
**Keywords:** Tsinghua University, chip breakthrough, photonic computing
**News Content:** Tsinghua University has made a significant breakthrough in chip research, opening up new possibilities for high-performance computing. According to the latest news on the 11th, Associate Professor Fang Lu from the Department of Electronic Engineering and Academician Dai Qiong Hai from the Department of Automation have jointly developed a distributed breadth photonic computing architecture. This breakthrough chip, named “Taichi” (Tai Chi), is an interferometric-diffractive heterogeneous integration chip designed to address the challenges of large-scale optoelectronic intelligent computing.
The “Taichi” chip departs from conventional electronic deep computing methods and innovatively employs photonic computing, achieving a general-purpose intelligent computing efficiency of 160 TOPS/W. This accomplishment is particularly noteworthy in the context of large-scale model通用 artificial intelligence. In an era where computational power demands are constantly increasing, this scientific breakthrough offers innovative inspiration, architecture, and pathways for enhancing high-performance computing capabilities.
This significant advancement has been published in the latest issue of the prestigious international journal ‘Science,’ signifying China’s leading position in photonic computing and a robust step forward in the global high-tech race. Tsinghua University’s innovative achievement is not only poised to drive progress in the research field but will also have a profound impact on future cloud computing, artificial intelligence, and big data processing industries.
【来源】https://www.cls.cn/detail/1644516
Views: 1