TigerGraph has released a cloud version of its graph database. TigerGraph Cloud supports AI and machine learning applications, and was announced at the annual Amazon Web Services Re:Invent conference. It follows a recent update of TigerGraph DB.

The developers of TigerGraph say the new cloud-based service will become available on other platforms in addition to Amazon Web Services, and that the service is a faster and easier way to run SQL-like queries securely in the cloud. TigerGraph was originally called SQLGraph.

Alongside the database, subscribers to the new service get starter kits for application development covering a number of areas such as anti-fraud and guarding against money laundering. Other kits include Customer 360,  and Enterprise Graph analytics. The kits include graph schemas, sample data, preloaded queries and a library of customizable graph algorithms - including PageRank, Shortest Path, and Community Detection.

TigerGraph was launched last year, and claims to offer the fastest graphics analytics implementation in its enterprise platform. The founder of TigerGraph, Yu Xu, led the massively parallel programming (MPP) team at Teradata before moving to Twitter. He then left to design a faster graph database that was natively distributed. Some other graph databases store data on top of Hadoop or NoSQL, though Neo4j, the current leader in the graph database market, stores its data in a graph structure with the data as nodes and the connections between data as edges.

Where TigerGraph differs is that it starts from a parallel model rather than adding it later, making more complex graph analytics possible alongside the ability to deal with much larger amounts of data. TigerGraph can run on more than 1,000 nodes of commodity Intel processors, meaning it can run queries with up to 20 hops without losing performance. Hops in graph databases are cases are equivalent to working through the connections between the data points, so a one hop query involves data points with a single connection, two hops move through an intermediate data point to find a connection and so on. The more hops, the worse the response time for queries. 

An updated version of the database was launched in September offering better integration with databases and storage systems, support for Docker and Kubernetes containers, availability on the Amazon Web Services Marketplace and Microsoft Azure, and a new graph algorithm library. Support for Spark is 'coming soon'.

The graph algorithm library is made up of GSQL implementations of graph analytics functions such as PageRank, Shortest Path, Connected Components and Community Detection. The library comes as a user-extensible set of GSQL queries. The developers describe GSQL as a Turing-complete, easy-to-use, SQL-like graph query language with built-in parallelism.

The company has also release a Neo4j migration toolkit that lets Neo4j developers transform Cypher queries into GSQL.

More Information

TigerGraph

Related Articles

RedisGraph Reaches General Availability

Kafka Graphs Framework Extends Kafka Streams

GraphLab Create

Open Source GraphQL Engine Launched

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info