Abstract:
In the dynamic landscape of data management, graph models and query
languages have emerged as powerful tools for modeling complex,
real-world processes. Among these, the property graph model and its
corresponding query languages, notably Cypher, have gained significant
traction in both industrial and research spheres. However, the
ever-growing demand for real-time data analysis presents a challenging
gap in these languages: their limited capability to handle streaming
graph data and continuous query evaluation.
Addressing this critical need, our latest work introduces SERAPH, an
innovative language built upon the foundations of Cypher, specifically
designed to support streaming features in property graph query
languages. SERAPH is not just an extension of existing paradigms; it
represents a paradigm shift. We have meticulously developed its
semantics by synergizing stream processing with the robust framework of
property graphs and time-varying relations. A key aspect of SERAPH is
its treatment of time as an integral component of its underlying
semantics, establishing a solid formal basis that is essential for
future practical implementations.
Furthermore, we delve into the syntax of SERAPH, carefully crafting it
to meet the needs of today’s industries. The talk will feature a
comprehensive demonstration of how SERAPH can be effectively utilized in
real-world scenarios, particularly focusing on graph-based continuous
queries that are rapidly becoming vital in various industrial applications.
Bio:
Riccardo Tommasini is a Maître de conférence (Associate Professor) at
INSA Lyon, LIRIS Lab and a Visiting Professor at the University of
Tartu, Estonia. Riccardo holds a Ph.D. from the Department of
Electronics and Information of the Politecnico di Milano. His thesis,
titled *Velocity on the Web*, investigates the velocity aspects that
concern the Web environment. His research interests span Stream
Processing, Knowledge Graphs, Database System Internals, Logics and
Reasoning, and Programming Languages. His Research has been recently
founded by the French National Research Agency ANR. Since 2015, Riccardo
has been attending and speaking at international events both in academia
(ISWC, ESWC, WWW, DEBS, EDBT) and industry (DockerCon, InfluxDays,
LinuxLab, Digit). He also spoke a numerous meetups like the Time Series
Meetup Tallin (200+ attendants), Kafka Meetup Milan (60+ attendants),
and Data Science Seminars Tartu (200+ online attendants).