Graph databases are exploding, thanks to the AI boom - here's why

2 hours ago 5
graph database concept
Cobalt88 / iStock / Getty Images Plus

Follow ZDNET: Add america arsenic a preferred source on Google.


ZDNET's cardinal takeaways

  • The graph database market, driven by AI, is increasing astatine a complaint of astir 25% annually.
  • Graph databases enactment cognition graphs, providing ocular guidance for AI development.
  • There are aggregate dedicated graph database vendors connected the market.

Over the past decade, determination has been endless churn successful technologies shaping the databases down the applications we run. The emergence of NoSQL databases, papers databases, and databases built connected the web for the web has brought america greater choices.

Also: Gartner says adhd AI agents ASAP - oregon else. Oh, and they're besides overhyped

More recently, determination has been a roar successful the usage of artificial intelligence (AI), some successful backend systems and done the emergence of generative technologies, creating an insatiable request for databases that tin grip and process super-sophisticated workloads. This request has led to the upsurge of graph databases and cognition graphs, which are ocular databases that tin assistance users negociate the requirements of AI. 

Graph databases person been connected the emergence for respective years and present comprise the fastest-growing class wrong the $137bn yearly database market. Again, convey AI -- graph databases are seen arsenic the astir optimal information backend for AI systems. Spending connected these technologies volition person a five-year compounded yearly maturation complaint of much than 26%, according to estimates published by tech expert Gartner astatine the extremity of 2024. The wide database absorption strategy marketplace volition turn 16% annually. In 2025, the Business Research Company projected a compound yearly maturation complaint of 24%.

Also: 95% of concern applications of AI person failed. Here's why

AI demands gobs of some structured and unstructured data, not lone fed into applications, but woven into connected patterns that present inferences. "The propulsion toward semantic knowing and reasoning successful AI systems is thing that level relational databases conflict to support," said Tony Tong, co-founder & CTO astatine Intellectia AI. 

Though abstracted and often confused, graph databases enactment manus successful manus with cognition graphs -- "a graph database is the tool, the motor for identifying connections wrong a fixed dataset. A cognition graph is the practice of the information itself -- the merchandise of a graph database," said Daniel Bukowski, main exertion serviceman astatine Data². 

Also: 71% of Americans fearfulness that AI volition enactment 'too galore radical retired of enactment permanently'

"Knowledge graphs supply AI systems with real-world accusation and however that accusation is related, which helps the AI reply questions with much accuracy and nuance. Graph databases let you to hunt done information much efficiently and supply discourse not recovered successful earthy information alone."

Graph environments tin beryllium applied to functions, specified arsenic real-time analytics, fraud analytics, retail, and logistics, said Shalvi Singh, laminitis of Healthengine.us, and elder merchandise manager astatine Amazon AI: "Knowledge graphs are aiding ample connection models by offering ample discourse for structured reasoning and by enabling contextual understanding." 

The ranking of the astir fashionable graph databases includes the pursuing technologies (source: DB-Engines):

  1. Neo4j Graph 
  2. Microsoft Azure Cosmos DB 
  3. Aerospike 
  4. ArangoDB
  5. OrientDB 
  6. GraphDB 
  7. Virtuoso 
  8. Amazon Neptune 
  9. Memgraph Graph
  10. NebulaGraph

Of course, implementing graph databases is not an overnight project. For example, "incorporating information from antithetic sources is inactive taxable to inconsistency oregon out-of-date information," Singh cautioned. 

Also: AI agents volition beryllium ambient, but not autonomous - what that means for us

Scalability is besides an issue, arsenic the show of these information environments whitethorn deteriorate arsenic datasets summation successful size and complexity. "These technologies bash not regenerate accepted databases," she added. More hybrid arrangements whitethorn beryllium indispensable for scalability purposes.

Plus, graph databases and cognition graphs "often necessitate specialized expertise, elaborate planning, and cautious structuring of interconnected data," said Bukowski. "Even though cognition graphs person been utilized for decades, graph databases are a newer, fast-growing conception of the database market, meaning that it tin beryllium hard to obtain, implement, and maestro some of these tools."

Also: Gen AI disillusionment looms, according to Gartner's 2025 Hype Cycle report

Without data, determination tin beryllium nary AI. For those looking to supply greater information enactment for their AI efforts, graph databases and adjoining cognition graphs correspond ocular connections that guarantee much on-target AI efforts.

Read Entire Article