A Cross-source Scheduling Method for Heterogeneous Data in Cloud Environment

Abstract

To address the time-consuming problem of scheduling the transmission of heterogeneous data across sources in cloud computing, many existing scheduling methods are implemented by heuristic algorithms, which usually cause load imbalance and low throughput and acceleration. Therefore, this paper proposes a cross-source scheduling method for heterogeneous data in a cloud environment, which carries out data prefetching before the actual scheduling, greatly reducing the computation amount during scheduling and thus the scheduling resource overhead. Then, all variables are updated, the quality of the heterogeneous data cross-source sub-stream to be scheduled is arranged, and it is regarded as the weight of the sub-stream data, the best quality sub-stream data among the heterogeneous multi-source sub-stream data is selected in the scheduling window each time for scheduling transmission, and the processing of all data sub-streams on paper is finished. The experimental results show that the method proposed in this paper is capable of cross-source scheduling of heterogeneous data in a cloud environment with high load balancing, throughput and acceleration ratios.