Follow Datanami:
March 25, 2015

Neo Tech Cranks Up Speed on Upgraded Neo4j

Graph database specialist Neo Technology has rolled out an updated version of its Neo4j tool that includes faster read and write performance capabilities aimed at critical graph database applications.

Neo4j 2.2 released on Wednesday (March 25) comes with souped-up write and read scalability, the company said. Expanded write capacity targets highly concurrent applications while leveraging available hardware via faster buffering of updates. A new unified transaction log is designed to serve both graphs and their indexes.

Graph databases are an advanced type of NoSQL database used for a variety of analytical and transactional tasks.

Neo Technology, San Mateo, Calif., also said the database engine update includes a new bulk import utility that can grab data from external sources at a sustained rate of 1 million documents per second to support graphs with tens of billions of nodes and relationships.

Meanwhile, upgraded read scalability includes a new in-memory graph cache capability to increase read throughput by up to ten times for highly concurrent transactional read applications, the company claimed.

Neo Technology said it has also added a new statistics-gathering capability to Neo4j along with a cost-based query optimizer for the company’s Cypher query language. The optimizer is touted as selecting the best query execution plan using built-in statistics containing data on graph size and shape. The upgrade runs up to 100 times faster “in certain cases” than earlier version of Neo4j, Neo Technology claims.

The graphic database upgrade also includes a batch of visualization and other tooling improvements aimed at boosting developer productivity. These include new graph visualization features, query plan visualizations and integrated tutorials and other training materials.

Neo4j 2.2’s fast write buffering architecture is designed to significantly improve write scaling, “both for initial loading of the graph and for highly concurrent transactional applications,” Neo Technology CTO Johan Svennson said in a statement announcing the latest Neoj4 release.

Graph database adoption continues to grow. Neo Technology cited market forecasts predicting that graph databases could be adopted by more than 25 percent of all enterprises by 2017. Graph analysis is increasingly being used in data-driven operations and for making strategic decisions after a data-capture design is in place.

Given the need for speed, Neo Technology is pitching the Neo4j upgrade as offering read and write performance as much a 100 times faster than previous versions.

Neo Technology announced a $20 million Series C funding round in January, which CEO Emil Eifrem said validates the graph database vendor’s earlier effort. While Eifrem is in a great position to boast about Neo4j’s success, the Swedish-born technology executive is more interested in promoting the values of graph databases as a whole. Graphs are about to break out, in a major way, he asserted.

According to recent rankings by DB-Engines.com, Neo Technology holds a comfortable lead in the graph database market.

Recent items:

Graph Databases Everywhere by 2020, Says Neo4j Chief

Forrester Ranks NoSQL Database Vendors

Datanami