Build your Data Mesh using Data Vault

Did you know that Data Mesh and Data Vault can work extremely well together? In this blog post, we’ll tell you how.

The Data Mesh hype

If you follow the world of data platforms, you must’ve been under a rock for a couple of years if you haven’t heard about Data Mesh.

Since the ideas were first published in 2019, the debate has grown over whether Data Mesh is the best way to get away from highly centralised data analytics infrastructures, and deliver the business insights required in a competitive business environment.

Predictions and questions

Whilst some industry experts and pundits are predicting the end for traditional data warehouse-based business intelligence and data analytics, the emergence of Data Mesh has also led some to question how Data Vault 2.0™ fits in this new world.

Data Mesh solutions

Is there a simple answer? A solution that takes the best of the Data Mesh‘s organisational and architectural approach and marries it with all the benefits of using the Data Vault method, while adhering to the best agile principles?

We’ve produced an easy-to-follow infographic that outlines how both Data Mesh and Data Vault complement one another to create an industry-leading data analytics and business intelligence function.

Data Mesh‘s chief exponent – Zhamak Dehghani – has admitted that there are many challenges that can only be confronted when you begin the process of constructing your Data Mesh solution.

We’d like to remind readers that there are no out-of-box vendor solutions for building a Data Mesh. However, embracing the Data Vault method as part of the solution gives access to plenty of proven options from very reputable providers with automation solutions.

Zhamak Dehgani identifies four key areas in a Data Mesh. We’ve identified some of the ways that Data Vault helps when implementing a Data Mesh in four key areas:

1. Data ownership

2. Data products

3. Federated governance

4. Self-service data platforms

Data ownership

Data ownership, within separate teams, closely aligned with a specific business domain, is at the heart of the Data Mesh philosophy. However, Data Vault modelling techniques allow each domain to be designed independently, without sacrificing important data integration capabilities.

Data products

Data Mesh stresses the need to create data products to manage and support user needs. Data Vault architecture includes a presentation layer which is managed as a set of individual data products.

Federated Governance

“Federated governance” has become synonymous with Data Mesh in the last couple of years, but what does it mean?

It can be defined as the process by which domain teams agree common principles of governance standards with minimal central management. They are also implemented locally.

Data Vault 2.0 has a set of standards that enables the implementation of common approaches. This facilitates data privacy or regulatory compliance and a full audit trail for data lineage.

Self-service Data Platforms

The great strength of Data Vault is its use of repeatable patterns and automation which helps support the creation of self-service data platforms.

Conclusion

We believe the Data Vault method can be used as a key element to underpin a Data Mesh implementation acting as a standard to help increase the compatibility in decentralised or federated architecture. Thus increasing the chances of success.