Graph database startup TigerGraph Inc. today announced a major update to its flagship cloud platform with the Savanna release, bringing with it six times faster network deployments and dozens of other new features.

The well-funded startup is one of the pioneers of graph databases alongside companies such as Aerospike Inc. and Neo4J Inc. Graph databases are designed to perform certain kinds of tasks more efficiently than traditional relational and not-so-traditional NoSQL database systems are capable of. For instance, there are many situations where databases need to store not only business records, but also contextual data about how those records are linked to one another.

As an example, a sales database might need to specify which store each sale took place in. With traditional Structured Query Language databases, doing this is difficult because their architecture doesn’t allow such connections to be discovered easily, making analytics a time-consuming and resource-intensive business.

But with TigerGraph and other graph databases, it’s possible to store these connections alongside traditional records, enabling rapid analysis and queries. It’s an extremely useful capability because certain kinds of applications, such as cybersecurity and artificial intelligence apps, can benefit from keeping tabs on these connections.

With the rise of AI and cybersecurity always a pressing threat, graph databases have become extremely popular, and so TigerGraph is under pressure to keep innovating to ensure its offering remains competitive.

The Savanna update is described as the “next stage in the evolution of the graph database.” According to TigerGraph, it enables massively parallel storage and computation to scale independently and without size limits, so it can keep up with the most demanding AI workloads.

Perhaps the biggest update pertains to deployments. TigerGraph says new compute resources can now be provisioned six times faster than any rival graph database platform. What’s more, these new deployments can be fully automated, configured and monitored via TigerGraph’s managed services.

Chief Executive Rajeevn Shrivastava said the company’s graph database was already the fastest and most scalable in the business even before today’s update. “Our Savanna cloud-native platform accelerates setup speed by six-fold while providing agility in how companies consume storage, compute and managed service options,” he said.

To help customers get started, TigerGraph has added nine preconfigured solution kits, which are essentially prebuilt versions of its graph database that are customized for different use cases. These include transaction fraud, application fraud, mule account detection, product recommendations, entity resolution, entity resolution KYC, customer 360, supply chain management and network infrastructure.

TigerGraph is also claiming that Savanna will deliver cost savings of at least 25% thanks to its new consumption model, in which customers only pay for what they use.

In addition, there are new, dedicated compute workspaces for online transaction processing and online analytical processing workloads, so customers can be sure that performance is optimized for each one, and it also supports three times as many data sources as before. With the update, TigerGraph adds support for high-speed data ingestion from Snowflake, Spark, Delta Lake, Iceberg and Postgres.

Other features include support for various query languages, including GSQL, OpenCypher and GQL, which is TigerGraph’s standard graph query language.

Finally, TigerGraph’s “bring your own cloud” policy means customers have the option to choose between a fully managed service in the cloud platform of their choice, or an on-premises deployment they manage themselves.

Image: SiliconANGLE/Meta AI