Yelp Open Dataset
A massive collection of businesses, reviews, and user interactions. We use this to test graph density and local clustering performance.
Amazon Product Data
Complete product metadata and reviews. Useful for testing path-finding between related items and massive indexing efficiency.
MovieLens 32M
Standard recommendation dataset. It helps evaluate bipartite graph traversals and collaborative filtering query speeds.
MovieLens Genome
Standard recommendation dataset. It helps evaluate bipartite graph traversals and collaborative filtering query speeds.
Stack Overflow
Interaction graph of developers. This dataset is used to benchmark community detection algorithms like Louvain or PageRank.
OnTime Flight Data
A network of flights between airports. We leverage this for time-series graph analysis and shortest-path routing performance.
FARS
NHTSA's nationwide census of fatal motor vehicle crashes. Crashes, vehicles, and people form an interconnected network for testing multi-hop traversals and aggregation over rich properties.
LDBC SNB
A synthetic social network from the Linked Data Benchmark Council at scale factor 100, with over 1.7 billion edges. We use its Business Intelligence workload to stress complex analytical queries at scale.