The LDBC Social Network Benchmark

The Social Network Benchmark's Business Intelligence workload (SNB BI) is a comprehensive graph OLAP benchmark targeting analytical data systems capable of supporting workloads. This paper marks the finalization almost decade research in academia and industry via Linked Data Benchmark Council (LDBC). SNB BI advances state-of-the art synthetic scalable database benchmarks many aspects. Its base sophisticated generator, implemented on distributed infrastructure, that produces social with small-world phenomena, whose value properties follow skewed correlated distributions where values correlate structure. temporal all nodes edges lifespan-based rules skew enabling realistic consistent inserts (recursive) deletes. query exploiting this correlation based LDBC's "choke point"-driven design methodology will entice technical scientific improvements future (graph) systems. includes first adoption "parameter curation" an benchmark, technique ensures stable runtimes variants across different parameter values. Two performance metrics characterize peak single-query (power) sustained concurrent throughput. To demonstrate portability we present experimental results relational DBMS. Note these do not constitute official LDBC Result - only audited can use trademarked term.