Distributed system resource optimal allocation method based on LSTM and genetic algorithm

The invention discloses a resource allocation method based on an LSTM (Long Short Term Memory) time prediction model and a genetic algorithm. The method comprises the following steps: 1) training a job execution time prediction model based on an LSTM network; 2) distributing a reasonable resource quantity for each job in batch jobs by using a genetic algorithm; a fitness function of the genetic algorithm is changed into a time prediction model based on LSTM, and the resource quantity suitable for each job is iterated through selection, intersection and variation of the genetic algorithm; 3) giving different resource quantities for different jobs by using a resource allocation algorithm based on a genetic algorithm; when the Spark distributed computing framework receives jobs, computing is carried out according to the quantity of cluster resources which can be used by different jobs, and the shortest processing time of the jobs is obtained. After the information of the batch jobs needing to be processed is sub