This paper considers how streaming real-time data applications can be programmed using the Real-Time Specification for Java, with the goal to leverage as much as possible the Java 8 Stream processing framework. Our approach is to buffer the incoming streaming data into micro batches which are then converted to collection for processing by the Java 8 infrastructure which is configured with a real-time ForkJoin thread pool. Our evaluation shows that there is a significant improvement in both response time and predictability when compared to using Java 8 alone.

BibTex Entry

@inproceedings{HaiTaoMe2016a,
 author = {HaiTao Me, Ian Gray and Andy Wellings},
 booktitle = {21st International Conference on Reliable Software Technologies -- Ada-Europe 2016},
 publisher = {Springer},
 title = {Real-Time Stream Processing in Java},
 year = {2016}
}