Bridging the Gap: CedarDB’s Unified Approach to Databases in the Age of AI
Databases have traditionally operated in two distinct realms: analytical databases designed to help businesses uncover trends and answer complex questions, and operational databases that manage the daily flow of transactional data such as orders and payments. However, CedarDB, a Munich-based database engine, is challenging this separation by supporting both functionalities within a single system. This innovation has gained particular relevance amid the rise of agentic artificial intelligence, as explained by CedarDB’s co-founder and CEO, Moritz Sichert, in a recent interview with TechCrunch.
For over a decade, CedarDB has been quietly evolving, building its foundation in academic research before gaining momentum in the enterprise sector. Many large companies have data infrastructures optimized for storing extensive transaction records but find themselves ill-equipped to rapidly answer analytical queries — especially those posed in natural language. CedarDB’s dual focus aims to address this gap efficiently without sacrificing speed or requiring extensive rewrites of existing applications.
From Academic Roots to Enterprise Readiness
CedarDB originated as a spinout from the Technical University of Munich, leveraging the research project Umbra. Umbra was designed from the ground up to outperform legacy database engines by maximizing the capabilities of modern hardware, emphasizing processing speed and efficiency. The CedarDB team, which includes five Umbra researchers among its founders, has maintained this high-performance ethos, promising faster database operations without increasing CPU, memory, or processing demands.
One strategic choice that sets CedarDB apart is its compatibility with PostgreSQL, a widely used open-source database system. This compatibility reduces the switching costs for enterprises, allowing them to adopt CedarDB without the need to rebuild existing applications. More intriguingly, it aligns with the logic that large language models (LLMs) are trained on, which enhances natural language querying capabilities and positions CedarDB well for AI-driven data interactions.
Balancing Innovation with Practicality
The journey from academia to production was not without challenges. As detailed in a blog post by co-founder Christian Winter, the team prioritized reliability, fault tolerance, and compatibility with popular tools — critical factors for enterprise adoption. The team also debated the timing of launching a cloud offering versus an on-premise product. Ultimately, CedarDB Enterprise was introduced as an on-premise solution alongside the free CedarDB Community Edition in May of last year, coinciding with a €5.3 million (~$6 million) seed funding round.
The cloud version was postponed but not abandoned. After beta testing with early adopters, CedarDB announced limited availability of its cloud offering at TechCrunch Disrupt, where it competes among 20 startups in the prestigious Startup Battlefield competition. This managed, cloud-hosted version is expected to broaden CedarDB’s appeal beyond large enterprises and better position it against competitors like Databricks.
Competing in the New Era of Data Architecture
Databricks recently introduced LTAP, a new data architecture unifying transactions, analytics, streaming, and operational data on a single storage copy, citing the dawn of the “agentic era” as a motivator. CedarDB’s CEO sees this as validation of the market’s need for integrated database solutions that can handle both transactional and analytical workloads efficiently.
Currently staffed by 14 employees, CedarDB plans to support its initial cloud users with the existing team but anticipates expanding as adoption grows. Sichert indicates that scaling up will coincide with a new funding round to accelerate growth and development.
Pictured from left to right: CedarDB founders Moritz Sichert, Lukas Vogel, Philipp Fent, Dominik Durner, and Christian Winter.
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.
For more detailed insights, read the full article Here.
