Skip to content
Kinetica As A Service - Now available on Azure. Learn More
SOLVE ROUTES & FIND RELATIONSHIPS AT SCALE

Graph Analytics

Bring together relational and graph analytics in one solution. Create graphs from relational data easily, find hidden relationships, optimize routes and supply chains, and seamlessly output graph analysis to tables for deeper analysis and dynamic graph data visualization. Drastically speed up and increase the breadth of your analysis.

DetroitGraph

HYBRID GRAPH DATASTORE

Kinetica graph is designed and implemented from the ground up to deliver ease of use and performance at scale. Kinetica’s graph analytics platform & database uses network agnostic and robust graph grammar for nodes, edges, and labels that makes it easy to translate columns into entities and relationships. Geographical coordinates can also be used to create nodes and edges from WKT points and linestrings. Kinetica’s graph database works in tandem with the distributed relational Kinetica database for I/O and seamlessly integrates at-scale OLAP expression support with hundreds of PostGIS-compatible ST_geometry functions.

PARALLEL AND DISTRIBUTED DATABASE ARCHITECTURE

Kinetica partitions graphs automatically and efficiently, distributing or replicating the graphs and solves without bottlenecks, where there is no need to process or store the entirety of the graph in one node. Convergence and overall time for path aggregation decrease in a scalable manner with the size of the cluster. Key differentiator of Kinetica Graph DB from its competitors is its efficient data representation supporting a very large number of edges/nodes that has no memory degradation under dynamic updates. Our optimized parallel Graph Solvers are built on top of this representation taking advantage of cluster computing for the best performance possible. See our list of solvers for more information.

DistributedGraph
MapMatching

MAP MATCHING

Kinetica’s patented Map Matching technology delivers unmatched accuracy and performance at-scale. Map Matching determines the roadways corresponding to noisy GPS data, using a novel adaptive width Hidden Markov Chain algorithm. See the article in the Journal of Geospatial Sciences for details.

SupplyChainOptimization

SUPPLY CHAIN OPTIMIZATION

Kinetica’s real-time multiple supply demand chain optimization (MSDO) routing solver enumerates millions of combinations in milliseconds and provides the dynamic routing and tracking for an entire distribution fleet to thousands of customer locations — easily administered from our graph UI web interface or via Kinetica’s Integrated Analytics SQL framework.

See our blog article on the MSDO solver.

ROUTE SOLVING

Generic, at-scale and at-pace parallel graph solvers, providing unmatched accuracy and performance operate in conjunction with a streaming relational database with solve-time restrictions and weights. The solvers are available via /solve/graph endpoint in Restful/R/C++/Python/Java/JavaScript API forms as low-level generic network solvers such as Dijkstra (shortest path), Travelling Salesman (with heuristics), Backhaul Routing, PageRank, MarkovChainProbability, Centrality Between-ness, Inverse Shortest Path, All Paths, etc.

See our technical documentation for the list of solvers Kinetica supports.

RouteSolving

ISOCHRONES

Kinetica’s isochrone feature helps you find accomodations with optimal distance to airports and business needs or find the best placement for your disaster recovery teams to optimize reaction times. Kinetica has the fastest isochrone solution on the market that integrates seamlessly with our distributed visualization engine. View isochrone maps directly in our product, or export images and vector layers (WKT) to be used with basemap providers.

Isochrones
propertygraphs

PROPERTY GRAPH QUERIES

Find hidden relationships in your data instantaneously. Kinetica’s powerful adjacency query engine is capable of traversing millions of graph nodes in many-to-many fashion with performance at scale. Our extendible and flexible graph grammar has feature parity with the Cypher query language.

Ready to Learn More?