Memgraph at Scale: Analyzing Company Ownership & Supply Networks With a 2 Billion-Node Graph

Sayari has built a global knowledge graph of corporate ownership and supply chains - with nearly 2 billion nodes and 7 billion edges.

To enable their users to run efficient traversals over this graph, Sayari experimented with several different graph databases before settling on Memgraph. In the process, Sayari identified some important lessons on efficiently storing large graphs in memory, scaling live analytic queries over millions of nodes, and modeling complex, real-world problems like beneficial ownership and international supply chain networks. Ultimately, Sayari ended up with a setup that combines an OLTP database's low latency with an OLAP database's processing scalability. In effect, Sayari uses Memgraph as a stand-alone graph analytics engine for their 2 billion-node knowledge graph.

What you will learn:

• How to efficiently store large graphs in memory
• How live analytical queries are unique to Memgraph
• How Sayari uses Memgraph’s hybrid in-memory architecture
• Why Sayari chose Memgraph over other graph databases