What is Event Sourcing?
Event Sourcing is an architectural pattern that is gaining popularity as a method for building modern systems. Unlike traditional databases which only store and update the current state of data, event-sourced systems store all changes as an immutable series of events in the order that they occurred and current state is derived from that event log.
Whenever an action occurs that changes the state of a business entity (for example an order is marked as shipped), the system will publish an event to the stream describing what has happened (as per the example, the event might be
OrderShipped). In EventStoreDB, additional metadata can also be stored in each event such as a timestamp, what action caused the event, and who performed that action.
The stream comprises a log of all events that have occurred, and by replaying them the current state can be derived. That can give the same end-result as in a traditional database, and much more; you can perform additional tasks such as time travelling through the system and root cause analysis. And being immutable it provides one of the strongest methods for audit log available.
Benefits of Event Sourcing
There are many benefits, not least being able to see a full history of state changes that is highly valuable in sectors like finance and government that require accurate audit logs from their systems and databases. Below we've listed the top benefits of event-sourced data models.
EventStoreDB stores your data as a series of immutable events over time, providing one of the strongest audit log options available (characteristics similar to a blockchain)
All state changes are kept, so it is possible to move systems backward and forwards in time which is extremely valuable for debugging and “what if” analysis
Root cause analysis
Business events can be tied back to originating events providing traceability and visibility of entire workflows from start to finish
Event-driven architecture: realtime streams
Unlike traditional databases which optimize pooling data into opaque silos, event streams can be listened to, allowing businesses to react in real-time to events
If a service goes down dependent services can “catch up” when the source comes back up - because events are stored in a sequence in the stream, knowing the current position allows synchronization
Replay and reshape
The series of changes (events) in a stream can be replayed and transformed as new business requirements inevitably emerge - e.g. the event stream can be replayed to a point in time and “what if” analysis can be used to project potential future outcomes
The migration of legacy systems to modern distributed architectures can be carried out in an incremental manner, gradually replacing specific pieces of functionality with event-sourced services (strangler pattern) - existing read paths of the legacy system can remain in place while writes are directed to the services
Event sourced systems strive for the minimum amount of synchronous interaction - consistency boundaries are consciously chosen so that business requirements are met and everything else is eventually consistent - this results in responsive, high performance, scalable systems
Since there is a log of all state changes of an application, it can be used in occasionally connected system scenarios - when a device is disconnected it can continue to work on its own data locally and synchronize upon connection
In event-sourced systems, events flow through queues and streams increasing observability - what is uniquely powerful is that the events can contain the business context which allows real-time observations and analytics
One way data flow
Data in a CQRS/event-sourced system flows one way (through independent models to update or read information) - this brings an improved ability to reason about data and debug as each component in the data flow has a single responsibility
Event streams are fundamentally just logs with strong backup and recovery characteristics - writing just the core “source of record” data to the event stream enables the rebuilding of the downstream projections - EventStoreDB itself is also a distributed database technology with failover if a leader fails
Why use EventStoreDB?
EventStoreDB is an industrial-strength Event Sourcing database that stores your critical data in streams of immutable events. It was built from the ground up for Event Sourcing - we believe that makes it the best solution in the market for building event-sourced systems.
It’s available as open-source software so it’s free to use. And with our tutorials it’s easy to get started - you’ll have a demo application to write and read events in minutes.
- Guaranteed writes
- High availability
- Multiple client interfaces
- Optimistic concurrency checks
- Subscribe to streams with competing consumers
- Great performance that scales
- Multiple hosting options
- Commercial support plans
- Immutable data store
- Atom subscriptions
Who uses Event Store?
Hundreds of companies around the world run production applications built on EventStoreDB. These companies depend on the power and immutability of data that Event Sourcing gives them.
It’s easy to get started with EventStoreDB.
If you are a new user, follow our getting started guide for step-by-step instructions on downloading and installing EventStoreDB, and writing sample applications to write and read events.
If you know the basics but want to know more, check out our tutorials for in-depth guides on specific features or use cases of EventStoreDB, our docs section for technical documentation, or head over to the community forum where we have an active community of users.