Application of Event Sourcing in Research Data Management

摘要

Event sourcing is an architecture pattern successfully applied in modern microservice-oriented web applications. It enables better scalability, integration, and traceability by changing the way in which data are handled in those distributed systems. There are many differences, however, between data used by commercial applications and research data. In this paper, we examine if and how event sourcing can be applied in the field of research data management and what ramifications and benefits can it bring. One of the most important rules of the pattern is to record and publish all the changes ever done to a data item. Therefore, not only the current version of the item exists, but also older versions and all modifications can be traced back in time. As we will show, it opens new avenues to work with research data. The publication of the changes makes it easy to replicate the data, and collaborate on them without a central authority. All these are features often required in the modern data-driven science. The concept, its suitability, ramifications, and initial performance evaluations are presented in two real world usage scenarios. The preliminary results corroborate the assertion of suitability of event sourcing in this particular field.