Skip to content
Platform Strategy

Introduction to Data-as-a-Product and Data Mesh

What is a Data Mesh?

A data mesh is a federated approach to data management for a company. A federated approach embraces the decentralization of data domains and their systems-of-record, while unifying discovery and access through a centralized governance layer. The concept of a Data Mesh is a natural evolution of the Data Pipeline.

Oracle says

“a data mesh solution should have some mix of data product thinking, decentralized data architecture, domain-oriented data ownership, distributed data-in-motion, self-service access and strong data governance.”

A data mesh is made up of:

  • Distributed data domains
  • Data discovery layer
  • Data governance

Distributed Data Domains

The distribution and federation of data domains across Trimble enables data stores to remain independent, performant and managed close to the products that they most immediately serve. Embracing data domain distribution also enables a more scalable, fault-tolerant architecture that lowers its dependence on a single point of failure. However, the integration of these various data sources to ensure the portability of data requires robust data governance that is centrally defined and enforced.

Data Discovery

In order to derive value from a data mesh, its contents must be discoverable and accessible. Trimble’s DataHub service accomplishes this by serving as a data marketplace and data pipelining service to surface and share datasets from disparate sources across Trimble. The service also provides mechanisms for consumption, primarily into analytical models and visualization tools. As we focus on this first phase of rolling out Trimble’s Data Mesh and Data Governance models, we are focusing on Trimble’s core enterprise digital transformation (DX) data assets.

Data Governance

This may be the most important aspect of a data mesh. Without central governance, the data mesh becomes an ugly, untrusted mess of data. This governance layer defines common consumer interface contracts, quality standards, security & privacy policies, and overall data model topology. Trimble’s specific Data Governance development efforts can be seen on the Trimble Cloud roadmap page.

What is Data-as-a-Product (DaaP)?

Data-as-a-Product describes a paradigm shift in the management of data from being a means to deliver specific customer value toward being a first-class business asset that is to be made available for general consumption for current or future business initiatives. It is part of Trimble Cloud’s mandate to establish a DaaP product culture and engineering approach within Trimble.

Zhamak Dehghani, the originator of the Data Mesh term, says “Data as a Product is one of the foundational pillars to move toward growing an innovation culture where data is readily and safely available for experimentation.”

Profiles

Profiles is a central entity-relation store that describes canonical data models and provides access to instances of “master data” assets. Profiles operates as a system-of-reference for the key attributes of entities such as Users, Applications, Devices, Organizations, and more.

DataHub

DataHub is a data marketplace and data pipelining service that is positioned to surface and share datasets from disparate sources across the Trimble ecosystem primarily for consumption via analytical models and visualization tools.

Supporting Services & Concepts

Trimble Cloud exemplifies the DaaP mindset through a variety of other supporting services and data artifacts:

  • Data Ocean - Trimble’s enterprise data lake

  • Trimble Resource Name (TRN) - Trimble’s universal resource identifiers

  • Managed Reference Data - Trimble’s directory of authoritative internal/external reference datasets

  • Domo - Trimble’s enterprise business intelligence, data visualization, and interrogation tool