Advanced Computing in the Age of AI | Thursday, February 22, 2024

AWS Launches New Analytics Engine That Combines the Power Of Vector Search And Graph Data 

One of the common debates in the AI circles is whether using graph or vector databases offers more truthful information in generative AI (GenAI) applications. While graph data is great at representing and analyzing complex relationships and connections, vector data is optimized for efficient search capabilities and calculations in high-dimensional spaces.  

Amazon Web Services (AWS) has decided to not debate this issue as it launched a new analytics database engine that combines the power of both capabilities. The general availability of the new service, named Amazon Neptune Analytics, was unveiled at the re-Invest conference in Las Vegas. 

Swami Sivasubramanian, vice president of data and machine learning at AWS, who announced the new service said “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”

Sivasubramanian further elaborated that the new service makes it easier for users to uncover hidden relationships across data – by storing the graph and vector data together. He also cited the example of Snap, one of the companies that use Neptune, who uses the service to find billions of connections among its 50 million active users “in just seconds”. 

The new service is available as a pay-as-you-go model with no one-time setup fees or recurring subscriptions. It is available now in some AWS regions including the US East, the US West,  Asia Pacific, and Europe. 

(Michael Vi/Shutterstock)

Since the launch of Neptune in 2018, it has become one of the leading services for storing graph data and performing updates and election on specific subside of the graph. However, one of the challenges has been that it takes some time to load the entire graph into memory. Loading large datasets from existing data lakes or databases to a graph analytic solution can take hours or even days. AWS Neptune Analytics addresses these issues by making the process significantly faster. 

AWS claims their internal benchmarking testing showed that Neptune is “80 times” faster than existing AWS solutions in finding insights in graph data and data lakes on S3. 

Amazon Neptune Analytics is a fully managed service, so AWS does all the infrastructure heavy lifting, enabling users to focus on workflows and problem-solving. The engine automatically allocates compute resources based on the size of the graph and quickly loads data in memory to run queries in seconds.  The service supports a library of optimized graph analytic algorithms and also facilitates the creation of graph applications using openCypher, one of the most widely used graph query languages. 

With the new capabilities,  Amazon Neptune Analytics could be a game changer in use cases that require quick response such as fraud detection and prevention, cybersecurity, and transportation logistics. 

Related Items 

Oracle Introduces Integrated Vector Database for Generative AI

DataStax Rolls Out Vector Search for Astra DB to Support Gen AI 

Retool’s State of AI Report Highlights the Rise of Vector Databases