The graph database vendor's platform now features improved workload management capabilities and support for Kubernetes to meet the data processing needs of large organizations.
TigerGraph on Wednesday unveiled its latest platform update, adding new capabilities designed to make the vendor's tools usable enterprise-wide rather than just for specialized applications.
Based in Redwood City, Calif., TigerGraph is a graph database specialist. The vendor's platform uses graph technology to enable customers to discover relationships between data points to then build data sets that can be used to inform models, dashboards and reports.
Unlike relational databases that enable a data point to connect to just one other at a time, graph databases enable data points to simultaneously connect to multiple others to create a neural network.
Those neural networks reduced the time it often takes to prepare data, thus speeding the decision-making process. In addition, those neural networks make graph technology ideal for specialized applications, such as fraud detection and social networking, that rely on discovering relationships between data points that might not obviously seem related.
However, graph databases such as TigerGraph, Neo4j and those offered by tech giants including AWS and Oracle are complex. For example, both TigerGraph and Neo4j require their own query language. That complexity limits their use by more than just those with deep expertise in computer science and statistics.
TigerGraph 3.9.3 aims to change that with features designed to make it more accessible and more scalable. It's the vendor's first platform update since founder Yu Xu was replaced as CEO by former senior vice president of engineering Mingxi Wu in May.
New capabilities
TigerGraph's update consists of new workload management capabilities, real-time data ingestion monitoring, support for Kubernetes and an integration with Neo4j query language OpenCypher.
The workload management capabilities automatically assign query tasks to the most available resources within TigerGraph to maximize productivity and reduce query wait times. Real-time data ingestion monitoring oversees the progress and accuracy of high-speed data loading. Support for Kubernetes enables TigerGraph to scale beyond specialized applications to handle enterprise-grade workloads.
Finally, the integration with OpenCypher -- now available after being unveiled in preview in 2022 -- enables users familiar with Neo4j's query language to use TigerGraph as well.
Overall, the update stands to benefit existing TigerGraph customers, according to Stephen Catanzano, an analyst at TechTarget's Enterprise Strategy Group. However, without the launch of a major new tool, the update likely won't attract new customers.
[It has] interesting additions, which likely have a positive impact on existing customers. But it's not a release that might drive new demand. It has nice feature enhancements.
Stephen CatanzanoAnalyst, Enterprise Strategy Group
"[It has] interesting additions, which likely have a positive impact on existing customers," Catanzano said. "But it's not a release that might drive new demand. It has nice feature enhancements."
In particular, Catanzano spotlighted the new workload management and real-time data ingestion monitoring as the most significant new capabilities for TigerGraph customers.
"Workload management helps to speed up the processing of data, and it ties into real-time data ingestion monitoring, which also ties into moving data faster," he said. "Organizations are all driving toward the delivery of real time to users for faster business decision-making, and these tie into that trend."
The theme that brings the release together is enterprise readiness, according to Victor Lee, TigerGraph's director of product marketing.
He noted that the workload management capabilities are designed to allow simultaneous workloads to run without delays. The real-time ingestion monitoring aims to enable organizations to ensure data quality across high volumes of real-time data, and the support for Kubernetes addresses the overall scale TigerGraph can manage.
"They're all useful for multiple teams using the system and internally managing the software," Lee said.
In addition, the new capabilities are designed to enable different personas within organizations to use TigerGraph, he continued.
For example, Kubernetes support appeals to system administrators seeking flexible deployment options, while the integration with OpenCypher appeals to developers writing applications for their databases.
GenAI in the works
Not included in TigerGraph's latest platform update are any generative AI capabilities, whether generally available or in preview. In addition, the vendor did not include generative AI in its only other platform update since OpenAI launched ChatGPT in November 2022, marking a significant improvement in generative AI and large language model technology.
Meanwhile, rival Neo4j in August improved its vector search and storage capabilities to help customers develop generative AI models. Numerous other database and data management vendors have also made generative AI a significant part of their product development plans.
Lee, however, said that TigerGraph does have generative AI capabilities in the works. He noted that TigerGraph in August both published a blog post and hosted a webinar on integrations between TigerGraph and LLMs.
"It's still in the development stage and not in the product yet, but there are a couple of things on the product roadmap for next year," Lee said.
One specific plan is to include natural language query capabilities that translate text to code as well as interpret the intent of the query so users don't have to phrase the query in a hyper-specific manner to get the desired response, according to Lee.
Another plan is to apply generative AI at the start of the graph database development cycle to provide suggestions as developers build a graph database.
Those capabilities, like many of the capabilities in the works from other database and data management vendors, stand to improve user productivity, according to Catanzano.
"Overall, [generative AI] is able to remove many manual processes, understand operator behavior and replicate the same behavior on similar tasks and overall increase productivity by eliminating manual and mundane tasks," he said.
He noted, however, that beyond just making data workers more productive by aiding them with generative AI, many data management vendors are adding vector search capabilities. The goal is to help customers discover the right data to inform their own development of generative AI and machine learning models.
"Generative AI is becoming big in databases, with a recent push on vector capabilities," Catanzano said. "I'm not seeing vector talked about by [TigerGraph]."
Future plans
Beyond generative AI, Lee said that TigerGraph's roadmap will continue to focus on making the vendor's platform more suitable for enterprise-wide use.
Included in that plan is greater emphasis on TigerGraph Cloud, which the vendor first released in 2019.
Catanzano, meanwhile, noted the importance of adding vector search capabilities as more organizations look to build their own data-driven AI and machine learning models.
Vectors help developers discover the data needed to train models, and the more data they can discover to inform a model the more accurate that model will be.
"Vector search is big to cross reference data and make recommendations of similar results," Catanzano said. "When you search on a home a set of parameters, vector searches can show you similar results."
Eric Avidon is a senior news writer for TechTarget Editorial and a journalist with more than 25 years of experience. He covers analytics and data management.
