Scheduling of fork-join tasks on multi-core processors to avoid communication conflict

The scheduling of fork-join model which is one of the most basic DAG models has a great impact on the efficiency of multi-core processors. Most of the task scheduling methods are based on the assumption of parallel communication. However, the communication in non-fully connected network is serial. In this case the regular methods will result in communication conflict, so that the number of processors and the schedule length will be increased. In order to solve this problem, we introduce a new task scheduling method TSFJ SC to map the fork-join tasks on multi-core processors. The number of processors can be strictly controlled in TSFJ SC. Busy-window is introduced to avoid the communication conflict. Strategies such as P0 insertion strategy, interval insertion strategy, task allocation strategy and so on are introduced to obtain the shorter schedule length. Through these methods, the TSFJ SC algorithm can obtain a better performance than other existing methods. The number of processors can be greatly reduced, the speedup can be greatly increased, and the efficiency of each processor can also be improved. Real-time processing can be obtained by TSFJ SC method.