The invention belongs to the field of
electronic information, and discloses a computing method based on
parallel computing power dynamic allocation under cloud edge fusion, which comprises the following steps of: in a pre-allocation stage, averagely dividing
system computing nodes into two groups, respectively adopting different allocation parameters, counting task arrival interval time in an
observation period, task
processing time of each group of nodes and
cloud computing processing time, and calculating the
cloud computing time according to the task arrival interval time, the task
processing time and the
cloud computing processing time; calculating the
completion time of each group of tasks under the corresponding distribution parameters, and performing parameter dynamic optimization configuration; in the stable operation stage, the calculation process is propelled by adopting the optimal distribution parameters; and simultaneously monitoring task arrival interval time, task processing time of each group of nodes and cloud computing processing time, if the variable change is relatively large, comparing
task completion time under the
current distribution parameter and the adjacent distribution parameter, judging whether to re-enter pre-processing, and performing parameter dynamic optimization configuration again. According to the method,
parallel computing power resources in the cloud edge fusion computing
system can be dynamically allocated, meanwhile, a
parallel computing algorithm is compatible, and the deduction
operation time of a
simulation platform is greatly shortened.