Transactions and Analytics. One Database.
CedarDB is the world's fastest database, PostgreSQL-compatible and built from the ground up to support OLTP and OLAP together. Run it free on your own infra, in your cloud, or fully managed.
Unified
No More Tradeoffs
CedarDB is a PostgreSQL-compatible, general-purpose database engine that handles transactional and analytical workloads on one engine and one copy of your data. Whether you're powering live dashboards, reports, or AI agents, CedarDB gets you fast answers from live data, without compromising between the analytical performance of data warehouses and the transactional guarantees you expect from a system of record.
The Tradeoff
Splitting transactions and analytics across Postgres and a separate analytics database adds pipelines, stale copies, failure modes, and another system to operate.
The CedarDB Approach
CedarDB runs transactions and analytics on the same data, so new writes are immediately available to dashboards and reports. Analytical performance no longer requires a separate engine, stale copy, or ETL pipeline.
The Tradeoff
Keeping analytics on PostgreSQL avoids stale copies, but long-running scans compete with application traffic and make interactive dashboards difficult to sustain.
The CedarDB Approach
CedarDB runs analytical queries alongside transactions, using modern parallel execution while preserving the ACID guarantees expected from a system of record.
The Tradeoff
Agent-generated SQL is complex and unpredictable. It can miss latency targets, overload the primary, or access more data than the task requires.
The CedarDB Approach
CedarDB answers complex queries quickly, while database roles and row-level policies constrain what each agent can read or change.
For Postgres Teams
Who is CedarDB for?
CedarDB is perfect for teams running Postgres whose dashboards, reports, or AI agents need answers from live data, without running a second database to get them.
You're in the right place if you either:
Run Postgres alongside a separate analytics database
Postgres handles your product's transactions, while a separate system, like Snowflake, ClickHouse, or Redshift, runs your dashboards and reports.
Have AI agents that need direct access to your data
Your AI agents connect straight to the database to answer questions, generate reports, summarize results, or update records on their own.
Run analytics queries against your production database
Your dashboards, reports, and ad hoc queries run against the same Postgres instance as your application, so they always work with the freshest data.
If one Postgres instance handles everything today and nobody's complaining, you don't need us yet.
And if you'd rather skip the ops entirely, a hosted CedarDB is on its way.
Modern on the Inside. Familiar on the Outside.
Built from Scratch
CedarDB shares none of PostgreSQL's code base, optimizing for today's hardware and workloads instead of carrying forward its pitfalls. No autovacuum falling behind on busy tables, no process-per-connection model capping how many clients you can serve, and full parallelism across every core for every query.
Read the architecture docs · Why it was time to start over
PostgreSQL Compatible
CedarDB implements the Postgres wire protocol, SQL grammar, and system catalog, so everything built for Postgres keeps working exactly as it does today. No plugins, no rewrites, no migration for the tools, drivers, and workflows you already rely on.
Read the compatibility docs · See the tools we have verified
Nine Years of Research
Before founding CedarDB, our team spent years developing Umbra at TU Munich and advancing the state of database architecture. That research now powers a production database built for fast transactions and analytics on modern hardware.
About the team · Amplify deep dive
Deploy Anywhere with Ease
Run CedarDB free on your own infrastructure with the Community Edition, add enterprise features and support for production with the Enterprise Edition, or let us run it for you with the fully managed CedarDB Cloud. Every path deploys in seconds with zero configuration and adapts automatically to wherever you run it.
Get started · Explore CedarDB Cloud
Engineered for Performance, from Start to Finish
See how CedarDB innovates at every step of query processing.
Cutting-edge innovation,
familiar interface
CedarDB speaks PostgreSQL SQL and integrates seamlessly with existing PostgreSQL tools. No learning curve, no friction.
CedarDB deploys in seconds and automatically adapts to any hardware, on premise or in the cloud.
Go beyond traditional SQL with built-in analytics for time series, JSON, vectors, and more.
Thorough analysis and optimization for every query
CedarDB's optimizer rebuilds your query from the ground up, so you don't have to trade query performance for ease of use.
CedarDB builds the optimal plan for each query using advanced statistics and transformations that materialize as little state as possible, as late as possible.
CedarDB's optimizer is also optimized to find the optimal plan for hundreds of joins on thousands of columns in milliseconds.
The speed of custom code
for every query.
CedarDB acts like your personal software engineer, generating virtually hand-optimized code for each query.
CedarDB intelligently chooses the best execution path for each workload, either generating assembly immediately or relying on LLVM for even faster code.
Maximize your
hardware’s potential.
CedarDB fully leverages the speed and capabilities of modern SSD and cloud storage, scaling to terabytes of storage and beyond.
CedarDB keeps the data you need close to the CPU, automatically switching between transaction- and analytics-optimized storage for each tuple.
CedarDB's scheduler adaptively allocates resources between queries, ensuring fair and fast execution of both short- and long-running tasks.
CedarDB, However You Need It
Let us manage it for you in the cloud, get enterprise support, or run it yourself.
Community Edition
Download and run CedarDB for free, on your own infrastructure.
- PostgreSQL Compatible
- Full ACID Transactions
- Advanced Analytics
- No Size Limits
- Role Based Access Control
- Monitoring
- Priority Support
- Management UI
- Automated Backups
- Agent Orchestration
Enterprise Edition
Get advanced features and support for running CedarDB in production.
- PostgreSQL Compatible
- Full ACID Transactions
- Advanced Analytics
- No Size Limits
- Role Based Access Control
- Monitoring
- Priority Support
- Management UI
- Automated Backups
- Agent Orchestration
CedarDB Cloud
Let us operate CedarDB for you, fully managed.
- PostgreSQL Compatible
- Full ACID Transactions
- Advanced Analytics
- No Size Limits
- Role Based Access Control
- Monitoring
- Priority Support
- Management UI
- Automated Backups
- Agent Orchestration
Want to have a first glance? Explore our demo of CedarDB Cloud.
Continue reading
Read more in the CedarDB Blog
Encoding or Compression: Why not both?
The terms "encoding" and "compression" are often used interchangeably, but they solve different problems and come with very different trade-offs. We break down what sets lightweight, data-aware encodings apart from general-purpose compressors like zstd, and when to reach for which.
CedarDB Launches HTAP (10 Years Ago)
Unifying transactions and analytics is once again making news headlines as the holy grail for AI agents. But it is hardly news. Umbra, the research behind CedarDB, started 10 years ago, and even then it was not the first. With CedarDB, what other vendors still call impossible is running in production today.
Prompts Don't Enforce Business Rules. Databases Do.
CedarDB co-founder Lukas Vogel on why AI agents are forcing a long-overdue conversation about where enforcement actually lives.