Gábor Szárnyas
Research Fellow, PhD in Software Engineering
- Address
- H-1117 Budapest, Magyar tudósok körútja 2., Building I, Room B421
- szarnyas [at] mit.bme.hu
- Phone
- +36 1 463 3598
- Homepage
- https://inf.mit.bme.hu/en/members/szarnyasg
I currently work at the Database Architectures research group of CWI in the Netherlands.
Research
My research interest is scalable graph pattern matching and benchmarking of graph query performance. I lead the research and development of the ingraph project, a live graph query engine that supports the openCypher language. I am also interested in the analysis of multiplex networks.
I am a member of the LDBC Social Network Benchmark and Graphalytics task forces, along with the openCypher Implementers Group. I am also an open-source contributor. I conducted research visits at the University of York (2015), McGill University (2016, 2017), the University of Waterloo (2017), VU Amsterdam (2019), and CWI (2020).
Talks
- GraphBLAS: A linear algebraic approach for high-performance graph algorithms, HPC developer room, FOSDEM (Brussels, 2020) and CEU, Department of Network and Data Science (Budapest, 2020). Blogpost on the latter.
- Business Intelligence workload: Progress report and roadmap, Linked Data Benchmark Council, 12th TUC meeting (Amsterdam, 2019)
- What makes graph queries difficult?, Neo4j meetup (Budapest, 2019)
- Multiplex graph analysis with GraphBLAS, FOSDEM (Brussels, 2019) and Neo4j meetup (Budapest, 2019)
- Mapping graph queries to PostgreSQL, Database meetup (Budapest, 2018)
- An early look at the LDBC Social Network Benchmark's Business Intelligence workload, GRADES-NDA at SIGMOD (Houston, TX, 2018)
- LDBC benchmarks: Three aspects of graph processing, Linked Data Benchmark Council, 11th TUC meeting (Austin, TX, 2018)
- Incremental View Maintenance for openCypher Queries, 4th openCypher Implementers Meeting (Copenhagen, 2018)
- Writing a Cypher query engine in Clojure, Clojure meetup (Budapest, 2018)
- Learning timed automata with Cypher, Neo4j meetup (Budapest, 2018)
- Graph-based analysis of JavaScript source code repositories, FOSDEM, Graph devroom (Brussels, 2018)
- Compiling openCypher graph queries with Spark Catalyst, Spark meetup (Montréal, 2017)
- Academic research on graph processing: Connecting recent findings to industrial technologies, openCypher Meetup (New York City, 2017)
- Social Network Benchmark: Business Intelligence workload, Linked Data Benchmark Council, 10th TUC meeting (Munich, 2017)
- ingraph: Live Queries on Graphs (slides), GraphConnect lightning talk (London, 2017) - also presented at the Neo4j Budapest Meetup
- The ingraph project and incremental evaluation of Cypher queries, 2nd openCypher Implementers Meeting (London, 2017)
- Incremental Graph Queries for Cypher, 1st openCypher Implementers Meeting (Walldorf, 2017)
- Incremental Graph Queries with openCypher, FOSDEM, Graph devroom (Brussels, 2017)
- The Train Benchmark: Cross-Technology Performance Evaluation of Continuous Model Queries, Linked Data Benchmark Council, 9th TUC meeting (Walldorf, 2017)
- Towards the Characterization of Realistic Models: Evaluation of Multidisciplinary Graph Metrics, MODELS conference (Saint-Malo, 2016)
Most important publications
- Saleem, M., Szárnyas, G., Conrads, F., Bukhari, S A C., Mehmood, Q., and Ngomo, A-C. N., "How Representative is a SPARQL Benchmark? An Analysis of RDF Triplestore Benchmarks", The Web Conference (WWW), 2019.
- Szárnyas, G., Prat-Pérez, A., Averbuch, A., Marton, J., Paradies, M., Kaufmann, M., Erling, O., Boncz, P., Haprian, V., and Antal, J. B., "An early look at the LDBC Social Network Benchmark's Business Intelligence workload", GRADES-NDA at SIGMOD, 2018.
- Marton, J., Szárnyas, G., and Varró, D., "Formalising openCypher Graph Queries in Relational Algebra", 21st European Conference on Advances in Databases and Information Systems, Springer LNCS, 2017.
- Szárnyas, G., Izsó, B., Ráth, I., and Varró, D., "The Train Benchmark: Cross-Technology Performance Evaluation of Continuous Model Validation", International Journal on Software and Systems Modeling, Springer, 2017.
- Szárnyas, G., Kővári, Z., Salánki, Á., and Varró, D., "Towards the Characterization of Realistic Models: Evaluation of Multidisciplinary Graph Metrics", ACM/IEEE 19th International Conference on Model Driven Engineering Languages and Systems, MODELS 2016.
- Szárnyas, G., Izsó, B., Ráth, I., Harmath, D., Bergmann, G., and Varró, D., "IncQuery-D: A Distributed Incremental Model Query Framework in the Cloud", ACM/IEEE 17th International Conference on Model Driven Engineering Languages and Systems, Springer LNCS, MODELS 2014.
Educational materials
- Cheat sheets for the technologies used in the research group
- Collection of openCypher publications
- Lecture notes on GitHub
- Prolog application for calculating functional dependencies
- Logic workshop
- Thesis templates: LaTeX, Markdown, Word
Student reports
-
2018
- János Benjamin Antal, Márton Elekes: Benchmarking Graph Information Systems, 1st prize & 1st prize on the national competition (co-advisor: József Marton)
- Petra Várhegyi: Matrix-Based Analysis of Multiplex Graphs, honourable mention
-
2016
- Dániel Stein: Graph-Based Source Code Analysis of Dynamically Typed Languages, 2nd prize (co-advisors: Dávid Honfi, Ádám Lippai)
- Gábor Szilágyi Gábor, Krisztián: Runtime Verification of Cyber-Physical System Using Graph Pattern Matching, 2nd prize (co-advisors: Márton Búr, András Vörös)
-
2015:
- Kővári Zsolt: Performance Analysis of Graph Queries, 1st prize (co-advisor: István Ráth)
- Maginecz János: Scalable Incremental Graph Query Evaluation, 1st prize & award on the national competition (co-advisors: Gábor Bergmann, István Ráth)
-
2014:
- József Makai: Optimization of Incremental Queries in the Cloud, 1st prize (co-advisors: Ákos Horváth, István Ráth)
- Dániel Stein: Incremental Static Analysis of Large Source Code Repositories, 1st prize & 1st prize on the national competition (co-advisor: István Ráth)
Awards
- Josef Heim Award for Outstanding Innovation Activities
- SIGMOD 2018 ACM Student Research Competition (sponsored by Microsoft) – 2nd prize
- Neo4j Featured Community Member
- Azure for Research Award
- MODELS 2016 ACM Student Research Competition (sponsored by Microsoft) – 1st prize
- Transformation Tool Contest (TTC) 2016 – 1st prize
- Schnell László Award (2015, 2016)
- STAF Doctoral Symposium – Best Presentation Award (2014)
- Pro Progressio Award for the report in the Scientific Students' Association Conference (2013)