Trovares Update 1.11 

Trovares has added functionality in two entirely new areas: connectors to databases/data repositories and Whole Graph Algorithms (WGAs) in its latest software release. See for yourself below. 

Trovares 1.11 - The Trovares Connector 

Trovares released an open-source Neo4j™ Connector and release 1.11.0 of its graph analytics engine on June 17, 2022. “Our Connector makes life better for Neo4j Cypher users,” said Jim Rottsolk, CEO.  “It allows Trovares xGT to plug directly into an existing Neo4j database and greatly accelerates the user’s Cypher analytical queries.  Faster results are important in mission-critical environments.”

Whole Graph Algorithms 

This new release also inaugurates Trovares’ catalog of whole graph algorithms (WGA).  “We chose to start with one of graph’s hardest algorithms, pattern search, or sub-graph isomorphism,” said David Haglin, CTO.  “Such patterns can be described easily in a Cypher or an SQL query but are difficult and may take forever to compute.  Our ability to do pattern search at scale in memory makes us a powerful tool in the compute layer of today’s modern data science software stack.

Page Rank & BFS

With this release, we are adding support for PageRank and Breadth-First Search (BFS) and we’ll be adding further whole graph algorithms (WGA) with subsequent releases starting with Connected Components.  These WGAs are commonly used by data science practitioners in ML/AI contexts”, Haglin continued.

Kerberos

Trovares 1.11.0 also has a feature where GSSAPI can be used to interact with a Kerberos authentication system.  “Our customers want to use our tool in Enterprise production environments where Kerberos support is required, said Rottsolk, adding: “We support full data access controls to an in-memory dynamic graph that can be Enterprise-wide in scale.  By concentrating on performance, we provide our customers with the fastest graph analytics available today without asking them to install another database.  We are an accelerator.  Our first connector integrates with Neo4j.  We are planning connectors for other applications in the future.”

In Summary 

  1. Trovares has begun a catalog of whole graph algorithms.  These are callable from a Cypher context using the CALL … YIELD syntax.  The First two are:   BFS and PageRank (along with Personalized PageRank).

  2. Support for Kerberos single sign on authentication is introduced in this release.

  3. Trovares now have a connector for integrating Trovares xGT and Neo4j that is released independently from Trovares xGT.  It can be found at PyPI.org under the name “trovares-connector”.

    1. Data formats (graph schemas) can be automatically inferred from a Neo4j database and created in xGT.

    2. Data movement between these two products has been optimized.