A method and apparatus for upgrading a process, a device and a storage medium. The method comprises: determining a target time period, wherein the target time period is obtained through prediction according to the distribution of the number of data processing requests received by a target process in a historical time period, and the number of data processing requests received by the target process in the target time period is less than a preset threshold, so that the target process may be upgraded within the target time period. Since it is predicted on the basis of a historical experience that it is likely in the future for other devices to not send data processing requests to a storage device within the target time period, when the target process is upgraded within the target time period, there is a greater possibility that the problem of data processing request congestion may be avoided, or even if the problem of data processing request congestion occurs, the number of congested data processing requests is usually relatively small.