Scheduling based on data validity in collaborative artificial intelligence of things

More and more researchers are paying attention to the research of industrial big data scheduling process in collaborative artificial intelligence of things (abbreviated as AIoT). However, there exists jitters in the process of data synchronization and interaction in the system, which leads to the failure of update transaction transmission, and the sensor data cannot be upload into the database. Therefore, The following key problems must can be considered. One is to reduce the runtime workload of transactions and data communication. The other one is to consider that when the transaction load demand of the system is distributed to different nodes, the tasks on each node need to communicate with each other during execution, and the jitters in the transmission cannot be ignored. The proposed method IMJB-EDF can solve the problems. The experimental results show that compared to the existing methods, the proposed method has better performance in transaction acceptance ratio, run-time load demand, and transaction aging.