MongoDB Acquires Firebolt
MongoDB has acquired Firebolt to embed a high-concurrency analytics engine directly into its database platform.
Decoder
- Data-movement tax: The operational cost, complexity, and latency penalty of copying data from operational databases into separate data warehouses for analysis.
- OLTP (Online Transaction Processing): Databases optimized for fast, transactional record updates.
- OLAP (Online Analytical Processing): Databases optimized for complex, read-heavy aggregate queries.
Original article
Real-time analytics on live operational data, built so applications and AI agents can act on what’s happening now.
The most valuable analytics no longer live in a dashboard. They live inside the application. A logistics app reroutes a shipment around a delay. A retail system reorders inventory before the shelf goes empty. A payments platform flags unusual behavior before the transaction clears.
Each of these decisions requires comparing live operational data against historical patterns. Which routes are backing up, and is that normal? What’s selling faster than usual, and should we restock? Is this login typical for this account at this time of day? Enterprises that can answer these questions in the moment adapt as conditions change. Those that can’t are reacting to yesterday.
Data infrastructure has long been split into two worlds: operational databases that run applications, and data warehouses that analyze what already happened. That paradigm was built for humans looking at morning dashboards, where data a few hours or days old was fine. As analytics moved into applications, real-time analytics emerged as a category of its own, combining complex queries over massive datasets with the speed, freshness, and concurrency that live applications demand. Agents raise the bar on all three. They act continuously, can drive thousands of concurrent queries, and need sub-second answers on fresh data. As businesses rely more on agents, real-time analytics becomes a core part of their data infrastructure.
Eliminate the data-movement tax
Running analytics on operational data often means copying it into a separate system, forcing organizations to build and maintain data pipelines, keep systems in sync, and run yet another platform. Even once the data is flowing, the added complexity, operational overhead, infrastructure cost, and risk of stale answers remain. This is the data-movement tax, and it compounds as agents drive more decisions.
Systems that require customers to move and synchronize data across separate platforms often struggle to deliver fresh answers at the speed and scale that live applications need. For MongoDB customers, the operational data their applications and agents rely on is already in MongoDB, so analytics should be able to run there too, at the source. That's why MongoDB acquired Firebolt.
One platform for operational data and real-time analytics
By integrating Firebolt into MongoDB’s Intelligent Data Platform, we’re bringing a high-performance analytics engine to where your operational data already lives. Firebolt’s engine will run right alongside your database, and we’ll keep it in sync automatically, so there’s no pipeline to build or manage. You’ll get one platform for operational and real-time analytical workloads, so teams can ship faster, with fewer systems to operate and secure.
Our Intelligent Data Platform already brings together operational data, search and vector search, Voyage AI models, stream processing, and the ability to build and run agents in production. Real-time analytics will extend that foundation, enabling applications and agents to analyze live data and act on it in the moment. Over time, customers will also be able to analyze data from other sources and formats alongside their MongoDB data.
When analytics powers a live application, an outage no longer just delays a report. It breaks the customer experience or stalls an agent mid-decision. MongoDB operates in this mission-critical world: today, approximately 75% of the Fortune 100 trust MongoDB with applications that require the highest levels of security, durability, and availability. We’re bringing that same level of reliability and operational excellence to real-time analytics.
Why Firebolt, and what’s next
Firebolt’s analytics engine is proven in production. It was built from the start for the high concurrency required by live applications and agents, not just internal dashboards and reports. Similarweb, for example, uses Firebolt to power complex, customer-facing analytics over more than a trillion rows at petabyte scale, with dynamic queries delivering sub-second performance without extensive pre-aggregation.
Firebolt’s team has tackled some of the hardest problems in analytical engines. CEO Benjamin Wagner’s background is in analytical database research at the Technical University of Munich. He built the Munich engineering team behind Firebolt’s engine and led engineering before becoming CEO. CTO Mosha Pasumansky was an architect of Microsoft Analysis Services, co-invented the MDX query language, and led the technical development of Dremel and BigQuery at Google.
“Firebolt was designed for the cloud to do what many people believed was impossible: low-latency, high-QPS analytics with fast-changing data on object storage,” says Mosha Pasumansky, Firebolt’s CTO. “We also knew that today’s applications deal with complex and flexible schemas, the kind MongoDB’s document model supports natively. So we made sure to keep the benefits of columnar storage and vectorized processing on those schemas through smart automatic column shredding.”
“Firebolt and MongoDB share the same DNA,” says Benjamin Wagner, Firebolt’s CEO. “We both build mission-critical data infrastructure for engineers. We spent the last several years building an analytics engine for what live applications and agents demand: high-concurrency, sub-second answers on fresh data. Joining MongoDB puts our engine where millions of developers already build. We can’t wait to bring real-time analytics to MongoDB customers.”
We’re thrilled to welcome the Firebolt team to MongoDB, and our teams are already working together to bring Firebolt’s engine into MongoDB. In the coming months, customers will be able to turn on real-time analytics for their MongoDB data, so their applications and agents can act on what’s happening now.
More to come soon. Join the waitlist to be among the first to try real-time analytics on MongoDB.
FAQ
What is Firebolt?
Firebolt is a high-performance analytical database designed for low-latency queries and high-concurrency applications.
Do Firebolt and real-time analytics replace a data warehouse?
No. Traditional data warehouses and data lakes remain the right fit for internal BI, reporting, and data science. Real-time analytics is for application-facing analytics, which has a different set of requirements: complex queries over massive datasets, sub-second responses, high concurrency, and fresh data for live applications and AI agents.
What happens to existing Firebolt customers?
Nothing changes for existing Firebolt customers. They can continue using the product with the same support and service. Firebolt’s customers are a big reason we wanted to bring this team and engine into MongoDB. They’ve shown what the technology can do under real production load.
Fresh dev news, AI-digested daily. Never miss a trend at devoured.fyi