Research on Multi-task Scheduling Algorithm for Integrated Energy Metering Acquisition System
Xiong Xia, Xiaofeng Tao, Xueliang Yang · IOP Conference Series Materials Science and Engineering · 2019
Abstract With the development of the energy Internet and the construction of a new energy measurement system, it is particularly important to promote the remote automatic collection and copying of water, gas, heat and electricity. Based on the electricity information collection system, this paper designs an integrated acquisition technology architecture system for electricity, water, gas and thermal energy measurement, and clarifies the technical route of the energy measurement system. Then, through the analysis of the system task classification, a multi-task adaptive real-time scheduling method based on Spark cluster is proposed, which solves the problem that the collection system has uneven task allocation, long overall task completion time and high priority based on the existing scheduling mechanism of the Spark cluster. The task timeout response causes problems such as poor user experience. Finally, the experimental results show that the total number of 2 million collection tasks are divided into six categories, and the default scheduling methods in the text and Spark are used for comparison. The time-based and priority-critical tasks are completed earlier than the default method. The response speed greatly improves the system user experience and effectively improves the level of social public services.