This blog was co-authored by Adrian Estala, VP, Data Mesh Consulting Services at Starburst.
We have been addressing data access challenges for a long time. For many years, the objective was to lock data down. We built layers of security around data to make access difficult; you had to prove you had a “right to access” before you were given the keys. Today, the goal is to create fast, simple, and secure paths from the data consumer to the data source. We are tearing down the complexity and driving democratization across our data consumers. For leading data-driven organizations, analytics teams must have automatic, instant access to the data they need. All of the compliance and risk controls are still very important, but we understand the incredible value that comes from unlocking data.
In our recent conversation on Data Mesh TV, we discuss the common pain points for obtaining access to data. We discuss how a Data Mesh approach can propel and elevate organizations to faster access while still maintaining strong governance. The mesh does not solve your access issues, but it does allow you to create visible Data Products that can be easily accessed and reused.
Who is making the data access decisions?
The business understands the value and risks associated with their data sets. The IT teams understand where the data sits and they manage the security gates along the data path. For the consumer, this feels more like a maze, with different entry and exit points. For any analytics solution, you are likely mapping out many different access paths depending on the source, the data classification and the consumer. Given the challenges of trying to access data from many different sources, many organizations headed down the disruptive centralization path. But, what if you could connect to all of those decentralized data sources via a standard Data Mesh design? One single point of entry to the Data Products that you need. In a mesh, you are focused on managing access to well-defined products.
Data Mesh establishes federated governance
In a standard mesh design, data sources are organized into Domains and those data sets are aggregated into reusable Data Products. With federated governance, the domain owner has the autonomy to manage access to the data from the connected data sources. You don’t need to chase data owners across different data sources, you work with a single domain owner. The domain owner will work with the data product owners to make access policy decisions.
Immuta enables access
Once the access decision is made on the front end, you need a secure and reliable technology to ensure that the access can be effectively executed. If you have a diverse set of data sources, enabling that access can be difficult. Immuta works across many different data sources and provides you a single pane of glass for managing access. The goal of the Data Mesh design is to simplify access on the front end, the consumer should be able to quickly find and access the data they need. On the back end, Immuta simplifies the configuration of the policies and rules for all of your data sources.