Distributed graph processing with Pregel and ArangoDB


Please note that the event will take place on October 5th 2017.

Most state of the art graph databases lack support for complex interactive graph algorithms. Some will require the use of external frameworks like Apache Spark, Hadoop or similar software which forces developers to export their data from the database. Developers who model their domain data in graphs can immensely profit by having the ability to execute custom algorithms right where the data is stored.

We integrated a pregel-style processing system into ArangoDB to allow the bulk-processing of distributed graphs with custom algorithms. Starting with ArangoDB 3.2 this new distributed graph processing system became available in our community edition. This allows you to reuse your existing ArangoDB clusters to run algorithms like PageRank, Vertex Centrality, Vertex Closeness, Connected Components and many more to gain analytical insights on your data, without having to use any external software.

In this online talk Simon, the leading mind behind our Pregel integration, will introduce the concept behind the Pregel computing model and show you how to use our implementation with your data.

The session will end with a short Q&A.

We'll be hosting this meetup on YouTube live:



• 9:00 PST (UTC - 8 hours)

• 12:00 EST (UTC - 5 hours)

• 16:00 UTC

• 17:00 BST (UTC + 1 hour)

• 18:00 CEST (UTC + 2 hour)