Answers are grounded in this paper and cite their sources.
Heterogeneity of data sources is a challenge for Big Data applications. This contribution presents an approach based on the conceptual foundation of the mediator/wrapper architecture where source data is transformed into a global schema and transferred to one or more data sinks with Apache Kafka. Given appropriate wrappers, arbitrary data sources and sinks can be connected with one another in a flexible fashion. The design and implementation of a prototype for a data set from a metagenomics study of are described.
No takes yet. Share an insight, caveat, or question.
Schmatz et al. (2018) studied this question.