Scaling Neo4j
Graph Database
Your Neo4j Graph Database provides unparalleled real-time data insights and deep understanding.
As its usage sprawls and scales new horizons, it needs to maintain the ability to rapidly navigate data relationships.
Why Graphs Need to Scale
When selecting a database, the platform’s ability to scale is crucial. How does the database handle large numbers of concurrent users, spikes in usage, and various geographies?
How Graphs Grow Very Large
A Variety of Graph Sizes
Here’s what different graph sizes may look like in practice as graphs grow. Operations and performance at scale become increasingly important:
Graphs of Things
Helps to visualize data and uncover questions to ask the database
Graphs of Transactions
Records official events against any of the “things”
Graph of Activity and/or Behavior
Records transactions, along with related activities
Neo4j’s scaling strategies
Autonomous Clustering
When throughput demand for the same dataset rises, Neo4j enables easy scale-out with Autonomous Clustering.
This architecture automatically allocates copies to the optimal servers based on default business rules or specified operational requirements.
Infinigraph
Data volume and sources inevitably grow over time, which creates a need to scale.
Infinigraph, the new distributed architecture, enables you to run operational and analytical workloads together in a single system at 100TB+ scale, without fragmenting the graph, duplicating infrastructure, or compromising performance.
A key capability of Infinigraph is property sharding, which helps you to scale out as your data grows. This technique keeps the structure of the graph in a single shard for performance and divides node and relationship properties into several shards running on Autonomous Clusters to achieve horizontal scalability for large graphs.
Fabric
If you have multiple graph databases for different sets of data, how do you treat your distributed graph as a cohesive whole?
The Neo4j Graph Database uses the Fabric architecture to bring these databases together. By quickly creating a composite database that federates a graph of all your remote graphs, you can perform a federated query across the entire graph.
Neo4j Ops Manager
As another side effect of scaling, today’s teams must cope with a growing number of scaled-out systems. They need a simple way to spin up new databases and administer, upgrade, and monitor them.
Neo4j Ops Manager empowers you to manage the Neo4j databases across all clusters and instances.
