@article{, author = {Kro{\ss}, Johannes and Krcmar, Helmut}, title = {Modeling and Simulating Apache Spark Streaming Applications}, journal = {Softwaretechnik-Trends}, series = {SSP 2016}, volume = {36}, number = {4}, year = {2016}, month = nov, abstract = {Stream processing systems are used to analyze big data streams with low latency. The performance in terms of response time and throughput is crucial to ensure all arriving data are processed in time. This depends on various factors such as the complexity of used algorithms and configurations of such distributed systems and applications. To ensure a desired system behavior, performance evaluations should be conducted to determine the throughput and required resources in advance. In this paper, we present an approach to predict the response time of Apache Spark Streaming applications by modeling and simulating them. In a preliminary controlled experiment, our simulation results suggest accurate prediction values for an upscaling scenario.}, issn = {ISSN 0720-8928}, url = {http://pi.informatik.uni-siegen.de/stt/36_4/01_Fachgruppenberichte/SSP2016/ssp-stt/22-Modeling_and_Simulating_Apache_Spark_Streaming_Applications.pdf}, }