Task Scheduling for Streaming Applications in a Cloud-Edge System.

With the increasing popularity of ubiquitous smart devices, more and more IoT (Internet of Things) data processing applications are deployed. Due to the inherent defects of traditional data transmission networks and the low latency requirement of applications, effective use of bandwidth computing resources to support the efficient deployment of applications has become a very important issue. In this paper, we focus on how to deploy multi-source streaming data processing applications in a cloud-edge collaborative computing network and pay attention to make the overall application data processing delay lower. We abstract the application into a form of streaming data processing, formalize it as a Stream Processing Task Scheduling Problem. We present an efficient algorithm to solve the above problem. Simulation experiments show that our approach can significantly reduce the end-to-end latency of applications compared to commonly used greedy algorithms.