Resilient scheduling of massive heterogeneous cloud resources considering energy consumption uncertainty

In the current elastic scheduling of massive heterogeneous cloud resources, the impact of energy consumption uncertainty is ignored, resulting in large errors in the scheduling results. To this end, we propose a study on elastic scheduling of massive heterogeneous cloud resources considering energy consumption uncertainty. Predicting the load of heterogeneous cloud resources. A three-tier scheduling model is constructed. Calculate scheduling errors considering energy consumption uncertainty and compensate for them. Elastic scheduling of massive heterogeneous cloud resources. Experiments show that the average scheduling task completion rate of the method reaches 97.85%, and the scheduling result error is small, which has high practical application value.