Remote sensing technology has played an important role in the emergency response of major natural disasters. However， the existing global emergency response mechanism based on remote sensing data still has problems such as insufficient data sharing， complex start-up procedures， and low response efficiency. It is urgent to establish an efficient， stable and sustainable remote sensing disaster emergency mechanism. This paper systematically reviews the value of multi-source remote sensing data in disaster emergency response and the shortcomings of the current international emergency response mechanism for major natural disasters. Based on the shared resource database of large-scale multi-source remote sensing data and the one-stop service collaboration method， the theoretical framework of China GEO Collaborative network of Disaster Data Response （CDDR） mechanism is proposed. It has also been applied in emergency response to disasters such as Tonga volcanic eruption and Turkey earthquake. Through two representative application cases， it can be seen that the mechanism has improved the efficiency of disaster emergency response services from various aspects such as data collection， download， analysis and application， and effectively supplemented the shortcomings of the existing mechanism. The new mechanism has simplified start-up procedures， enhanced data aggregation capabilities， more professional disaster assessment capabilities， and more accurate sharing capabilities， which is expected to provide stable and sustainable sharing services for the international community.