Precipitation is an intermittent event; the higher the spatiotemporal resolution of precipitation data, the better it can characterize the processes of extreme precipitation events. In this study, we analyzed the spatiotemporal evolution of the global annual maximum 3-hr precipitation processes from 2000 to 2024 and its influencing factors based on high temporal resolution precipitation data from Multi-Source Weighted-Ensemble Precipitation (MSWEP). The results show that from 2000 to 2024, the annual maximum 3-hr precipitation exhibited a significant increasing trend over 6.66% of the global land area. In arid regions, extreme precipitation events are more concentrated within the peak 3-hr window, whereas in humid regions, a greater ratio of total event precipitation is distributed in the hr before and after the peak. The timing of the annual maximum 3-hr precipitation differs hemispherically, primarily concentrated during the local summer; these events also exhibit differences in their preferred time of occurrence within a day. Over the past two decades, precipitation return values have increased across more than half of the global land area. Among the climatic drivers of extreme precipitation, Land-Ocean Temperature (LOT) exerts the strongest influence, followed by the Nino 3.4. This study can help deepen the understanding of the evolution processes of global sub-daily extreme precipitation and their influencing factors.