Dynamic priority scheduling of periodic queries in on-demand data dissemination systems

作者:

Highlights:

• We propose EDFS, a variant of the classic EDF scheduling algorithm, which can be used to schedule real-time tasks in the data broadcast environment. We also provide a necessary and sufficient schedulability test for EDFS.

• Based on EDFS, we propose, to our best knowledge, the first dynamic priority based broadcast scheduling algorithm, EDFS-BS, which comprehensively considers the real-time characteristics of the tasks, the sharing feature of the broadcast data items and the continuity of the broadcast services. We also analyze the schedulability of EDFS-BS and provide a bandwidth utilization based schedulability test for it.

• We conduct extensive experiments to evaluate the performance of EDFS-BS versus existing solutions with comparable quality. The experimental results reveal the efficiency of EDFS-BS as compared to its competitors, in terms of service ratio and bandwidth consumption.

摘要

•We propose EDFS, a variant of the classic EDF scheduling algorithm, which can be used to schedule real-time tasks in the data broadcast environment. We also provide a necessary and sufficient schedulability test for EDFS.•Based on EDFS, we propose, to our best knowledge, the first dynamic priority based broadcast scheduling algorithm, EDFS-BS, which comprehensively considers the real-time characteristics of the tasks, the sharing feature of the broadcast data items and the continuity of the broadcast services. We also analyze the schedulability of EDFS-BS and provide a bandwidth utilization based schedulability test for it.•We conduct extensive experiments to evaluate the performance of EDFS-BS versus existing solutions with comparable quality. The experimental results reveal the efficiency of EDFS-BS as compared to its competitors, in terms of service ratio and bandwidth consumption.

论文关键词:On-demand,Data dissemination,Bandwidth consumption,Dynamic priority,Periodic queries

论文评审过程:Received 15 July 2016, Revised 19 January 2017, Accepted 22 March 2017, Available online 27 March 2017, Version of Record 28 March 2017.

论文官网地址:https://doi.org/10.1016/j.is.2017.03.005