A dynamically adaptive distributed system for processing complex continuous queries
Bin Liu, Yali Zhu, Mariana Jbantova, Bradley Momberger, Elke Angelika Rundensteiner · 2005
Recent years have witnessed rapidly growing research attention on continuous query processing over streams [2, 3]. A continuous query system can easily run out of resources in case of large amount of input stream data. Distributed continuous query processing over a shared nothing architecture, i.e., a cluster of machines, has been recognized as a scalable method to solve this problem [2, 8, 9]. Due to the lack of initial cost information and the fluctuating nature of the streaming data, uneven workload among machines may occur and this may impair the benefits of distributed processing. Thus dynamic adaptation techniques are crucial for a distributed continuous query system.