Graph processing systems back to the future

Abstract

Graphs are data model abstractions that are becoming pervasive in several real-life applications and use cases. In these settings, users focus on entities and their relationships, further enhanced with multiple labels and properties to form the so called property graphs. Modern graph processing systems need to keep pace with the increasing fundamental requirements of these applications and to tackle unforeseen challenges. Motivated by a community vision on future graph processing systems [6], in this talk I will present the system challenges that are lying behind the current research topics on graph processing and graph analytics. Many current graph query engines support subsets of graph queries that they can efficiently evaluate, thus disregarding more expressive query fragments on top of property graphs [2]. It becomes crucial to address efficient query evaluation for complex graph queries, as well the extensibility of the underlying graph query and constraint languages [1, 3]. Moreover, the dynamic aspects [5] of evaluating queries on streaming graphs are equally important and need to be considered in ongoing and future benchmarking efforts [4]. The overarching goal of my talk is to touch upon our past and ongoing work on these topics and to pinpoint the research directions shaping the already bright future of graph processing systems.