TigerGraph Cloud adds graph analytics, machine learning tools

Graph database provider TigerGraph on Tuesday said that it was adding graph analytics and machine learning tools to its graph database-as-a-service (DBaaS) TigerGraph Cloud,  which was was launched in 2019 and is available across major cloud platforms such as Amazon Web Services (AWS), Microsoft Azure and Google Cloud (GCP).

Dubbed TigerGraph Insights, the visual analytics feature will come in the form of a no-code, low-code tool to allow nontechnical users to create visual representations of business insights on top of the database platform via dashboards, the company said.

“The Insights tool connects intuitive graph data with traditional business intelligence to produce multidimensional and interactive graphics. These graphics can also be connected within an interactive dashboard application for easy sharing to gain deeper understanding and insight into connected data,” said Jay Yu, vice president of product and innovation at TigerGraph.

Prior to offering the low-code Insights tool, the company provided connectors for analytics tools such as Microsoft Power BI, Tableau and Google’s Looker.

“The challenge for those business insights (BI) tools is that we have to do a reverse transformation from a graph connected model back to the relational model, because those tools are relational in nature,” Yu said, adding that these connectors didn’t offer the ability to see the connected graph data.

The other way to circumvent the issue before the launch of TigerGraph Insights was to upload the data into a development tool, dubbed Graph Explorer, and run simple queries, Yu explained, adding that this approach would be time consuming.

TigerGraph Insights, which was made available for on-premises users around three months back, has now been made generally available for managed service users, the company said.

Python development framework embedded within Jupyter notebook

In order to provide data scientists with the ability to quickly build artificial intelligence or machine learning models using connected graph data, the company also is launching a Python development framework, dubbed ML Workbench, which offers integration with the open-source data science toolkit Jupyter Notebook.

ML Workbench—which is in general availability and can also be used with Amazon SageMaker, Google Vertex AI, and Microsoft Azure ML—can improve machine learning model accuracy, shorten development cycles thereby delivering more business value, the company said, adding that data scientists can also take advantage of TigerGraph’s massive parallel graph data compute engine and over 55 open-source graph algorithms.

Both ML Workbench and TigerGraph Insights, along with the likes of the company’s Graph Studio, GSQL Shell, Graph QL Gateway and AdminPortal, can be accessed via the TigerGraph Suite portal.

TigerGraph, which is offering the new tools under its free-tier service, said that enterprise pricing was dependent on compute demand and usage.

The company, which says it’s seeing uptake in the financial services, healthcare and gaming sectors along with use cases in supply chain management, said that most of its paid customers were still operating on-premises. However, Yu said that the company is gaining more cloud-based users.