LDBC Benchmarks are a great source of tests that really hone in on the things we need to benchmark:
LDBC Social Network Benchmark (LDBC-SNB)
The Social Network Benchmark's Interactive workload is focusing on transactional graph processing with complex read queries that access the neighbourhood of a given node in the graph and update operations that continuously insert new data in the graph
The Social Network Benchmark's Business Intelligence workload is focusing on aggregation- and join-heavy complex queries touching a large portion of the graph with microbatches of insert/delete operations
LDBC Graphalytics Benchmark (LDBC Graphalytics), focuses on large-scale graph analysis
LDBC Semantic Publishing Benchmark (LDBC-SPB) SPB performance is measured by producing a workload of CRUD (Create, Read, Update, Delete) operations which are executed simultaneously. The benchmark offers a data generator that uses real reference data to produce datasets of various sizes and tests the scalability aspect of RDF systems. The benchmark workload consists of (a) editorial operations that add new data, alter or delete existing (b) aggregation operations that retrieve content according to various criteria
These are a really good mix of both static and streaming datasets, which is exactly what we need.
Prior art from Frank McSherry w/ DDFlow
LDBC Benchmarks are a great source of tests that really hone in on the things we need to benchmark:
These are a really good mix of both static and streaming datasets, which is exactly what we need. Prior art from Frank McSherry w/ DDFlow