Data Mesh
The data mesh is a decentralized architecture and governance concept that puts the responsibility for data on the teams that produce—and actually own—the data. There will be no centralized ownership of data. Instead, each team will monitor their data, analyze, and transform it.
For example:
Financial team members would be responsible for the data in the financial domain.
Despite its decentralized nature, data mesh relies on centralized governance principles and self-service infrastructure to ensure that data is usable (data products) across the entire organization regardless of its department. This way, it distributes responsibility without the possibility of creating data silos.
There are four pillars of data mesh (if you’d like to read more about them in detail, check our article here):
- Domain-oriented decentralization
- Data as a product
- Self-service data infrastructure
- Federated governance
Data mesh is a relatively new concept in the data landscape, but it’s gaining popularity because of the many benefits it can bring. A successful data mesh architecture can produce more motivated and efficient teams, accelerate the delivery of data products, and is reliable and extendable as your organization expands.
However, data mesh isn’t for everyone. You will still have to consider the possibility of governance debt and potentially recreating a data silo environment. Regardless of your intention for the data mesh, any data-driven organization needs to be familiar with new data architecture concepts as they emerge.
