SingleStore boosts database performance to meet AI needs
With AI development and deployment imposing greater workload demands on data management systems than traditional analytics, the vendor's latest update targets efficiency at scale.
SingleStore on Tuesday unveiled a spate of new capabilities for its database platform, including performance at scale improvements aimed at better enabling enterprises to develop AI applications.
In addition, new features target data ingestion and integration, AI development on data lakehouses using an updated integration with Apache Iceberg, AI deployment and the developer experience.
Collectively, SingleStore's update advances its database's capacity for AI development, according to William McKnight, president of McKnight Consulting Group. Most significant, however, is the boost to performance at the scale demanded by AI workloads, which are larger than those of analytics workloads, he noted.
"The performance enhancements are crucial for AI applications that require rapid data processing, efficient querying and scalable infrastructure," McKnight said. "While other features … are also important for AI deployment, the performance improvements have the broadest impact on the overall scalability and efficiency of AI workloads."
The performance enhancements are crucial for AI applications that require rapid data processing, efficient querying and scalable infrastructure. While other features … are also important for AI deployment, the performance improvements have the broadest impact on the overall scalability and efficiency of AI workloads in the enterprise today.
William McKnightPresident, McKnight Consulting Group
Kevin Petrie, an analyst at BARC U.S., similarly noted the value of the update. But given that it largely improves existing capabilities rather than adds new ones, he said it is gradual.
"These enhancements are incremental improvements to SingleStore's market advantages in terms of performance and the ability to support operational and analytical workloads concurrently," he said.
Based in San Francisco, SingleStore is a database vendor whose platform is designed to enable customers to quickly ingest data from a wide array of sources to inform near-real-time decisions. Like SingleStore, Aerospike, Neo4j and Oracle have all recently prioritized database performance.
New capabilities
With many enterprises now using proprietary data stored with vendors such as SingleStore to build AI applications that understand their unique operations, numerous data management vendors have created environments that simplify AI development.
Databricks unveiled new capabilities last week aimed at simplifying agentic AI development and building the data pipelines that feed AI models and applications. Similarly, Snowflake on June 3 introduced a conversational interface powered by agentic AI for data management and analysis and data integration capabilities aimed at using structured and unstructured data together.
Features SingleStore has added to enable AI development with its database include an initial integration with Apache Iceberg and improved vector search and storage capabilities. In addition, in October 2024, the vendor acquired BryteFlow to better enable users to ingest and integrate data from varied sources.
SingleStore's latest additions addressing database performance were motivated by a combination of customer feedback and the vendor's own long-term product vision, according to Dave Eyler, vice president of product management at the company.
"No one ever asks for a slower database … so we're constantly innovating to ensure SingleStore remains [fast] even as workloads grow more complex and data volumes surge," he said.
Specific features of SingleStore's latest database platform update include the following:
Tools targeting improved database performance, such as automatic query optimization and multi-value indexing for JSON.
The use of SingleStore Flow -- the ingestion and integration tool resulting from the acquisition of BryteFlow -- within the vendor's Helios database-as-a-service platform to simplify moving data from Snowflake, Postgres, SQL Server, Oracle and MySQL.
Hosting Cloud Functions -- serverless functions that can be used to build applications such as agents and APIs -- in Aura Container Service, which is SingleStore's serverless compute platform.
A new speed layer on Apache Iceberg to make it faster to work with Iceberg tables.
Features attempting to simplify the developer experience using SingleStore's database, including an integration with GitHub, easier pipeline monitoring capabilities and a SQL Editor with autocomplete.
AI development depends on synchronizing data, model and application lifecycles, according to Petrie. As a result, he highlighted the significance of Aura Container Service and the integration with GitHub.
"The Aura Container Service is a good move because it helps build and integrate workflows or tools in a modular fashion on the SingleStore platform," Petrie said. "SingleStore is also wise to integrate with GitHub and notebooks to assist AI developers."
Regarding how SingleStore's database capabilities compare to those of its peers, Petrie added that SingleStore stands out by integrating transactional and analytical processing to support real-time analytics and AI-powered tools.
"Gorillas such as Google, Oracle and AWS will continue to dominate the database market, but SingleStore has carved out a differentiated niche," he said.
McKnight likewise called out the significance of hosting Cloud Functions in Aura Container Service, in addition to addressing performance at scale.
"These features address critical pain points for users working with large-scale AI applications and complex data sets, enabling faster data processing and more efficient model deployment," he said.
"SingleStore is … a leading and innovative vendor that is significantly pushing the data management market forward, particularly in real-time data processing and enterprise AI," he said. "SingleStore's commitment to continuous improvement and developer experience further solidifies its position as an innovator, aiming to remove limits and barriers for builders deploying AI on their data."
Next steps
As SingleStore plots product development for the rest of 2025, making AI more accessible, powerful and integrated with data is the vendor's focus, according to Eyler.
Features such as a no-code for querying and analyzing data and Agent Studio, an environment for agentic AI development natively on SingleStore, are part of that initiative, he said.
McKnight suggested that SingleStore could potentially better serve its customers and attract new ones by expanding beyond its database roots to venture into other areas of data management.
"SingleStore could provide more comprehensive monitoring and observability tools, more advanced analytics and visualization capabilities, and broader cloud platform support," he said.
Petrie, meanwhile, noted that more integrations with AI and machine learning providers would be a way for SingleStore to improve its AI development and management capabilities, especially as agentic AI evolves.
"I'd recommend that SingleStore extend its AI ecosystem by partnering with AI/ML platforms such as Dataiku, as well as vendors that are rolling out agentic tools and applications for business functions," he said.
Eric Avidon is a senior news writer for Informa TechTarget and a journalist with more than 25 years of experience. He covers analytics and data management.
