With Neptune Analytics, AWS combines the power of vector search and graph data

There’s been a debate of sorts in AI circles about which database is more important in finding truthful information in generative AI applications: graph or vector databases. AWS decided to leave the debate to others by combining the best of both capabilities in a new service announced today at AWS re:Invent called Amazon Neptune Analytics.

Swami Sivasubramanian, vice president of data and machine learning at AWS, announced the new tool onstage at the AI keynote at re:Invent, saying that the cloud giant wanted to create something that combines the best of both approaches. “Since both graph analytics and vectors are all about uncovering the hidden relationships across our data, we thought to ourselves: ‘what if we combined vector search with the ability to analyze massive amounts of graph data in just seconds,’ and today, we are doing just that,” he said.

He explained that the new service helps customers analyze existing Neptune graph data or data lakes on top of S3 storage, taking advantage of vector search to find key insights. “Neptune analytics makes it easier for you to discover relationships in your graph with vector search by storing your graph and vector data together,” he said.

Ray Wang, founder and principal analyst at Constellation Research, says the new tool is a clever combining of the two technologies. “Vector databases are key for managing and querying high-dimensional data for machine learning. Graph is awesome for relationship mapping,” he told TechCrunch. “Vectors get bogged down in dimensionality, while graphs are hard to scale. Traditionally, these two are separate. Putting them together addresses the weaknesses of the other.”

Neptune Analytics is a fully managed service, meaning AWS “takes care of the infrastructure heavy lifting, enabling you to concentrate on problem-solving through queries and workflows. Neptune Analytics automatically allocates compute resources according to the graph’s size and quickly loads all the data in memory to run your queries in seconds,” the company wrote in a blog post announcing the new feature.

Neptune Analytics is available starting today as a pay-as-you-go service in seven AWS regions, including US East (Ohio, N. Virginia), US West (Oregon), Asia Pacific (Singapore, Tokyo) and Europe (Frankfurt, Ireland).