Data Mesh in Action (Data Mesh in action) (Data column)

I have read this book by three authors Maikhzhak, Balnoyan and Sivak over the past month. The book itself can serve as a simple guide to understanding and implementing the data mesh concept. (or data grids as indicated in the translation)It decentralizes data management to improve scalability, availability, and extract value from data. The book consists of three parts and nine chapters.

Part 1: Foundations This part is devoted to the basics and allows you to easily and quickly understand what the whole Tsimes is. 1. The What and Why of the Data Mesh In this chapter, the authors describe how the paradigm of working with data has changed from centralized DWH to decentralized ownership and operations, which should be behind the teams responsible for the domains themselves. The point is to look at the entire sociotechnical system and focus on people and processes, not just technology. The goal is to remove the bottlenecks of a centralized system. In fact, this is similar to the story of microservices in general, and also the fact that the effect of the grid is only visible on a large scale. There is a chapter about this in Vlad Hononov’s book “Learning DDD”, which I examined in the article.DDD and data mesh" 2. Is a Data Mesh Right for You? The second chapter allows you to understand how data mesh is suitable for your needs. The advantages and disadvantages of a centralized and decentralized approach are discussed here. We talked about this a lot with Kolov in the head. "Research Insights Made Simple" 3. Kickstart Your Data Mesh MVP in a Month In this chapter, the authors forge the iron while hot, or rather tell how to saw down the MVP for a data grid for a month. They coolly explain how to deal with stakeholders, choose the right domain, and then from shit and sticks to assemble a prototype that will show the effect.

Part 2: The Four Principles in Practice Here, the authors analyze in detail all four basic principles of data mesh, as bequeathed by Zhamak Dehghani, the author of the concept itself. 4. Domain Ownership Here we are talking about responsibility for the domain, when the data begins to be responsible for the team, which is closest to their generation. It is important to correctly define the boundaries of data domains so that they are aligned with business functionality and help the interaction of technical and business stakeholders. 5. Data as a Product This chapter describes the principle that data becomes a first-class citizen among products, not a side effect. This change leads to the fact that the product date has properties: easy detection, reliability, the ability to use by consumers. To do this, you need to follow a number of practices: the presence of clear APIs, metadata, quality standards, and so on. 6. Federated Computational Governance This principle is about a decentralized federal system for computing. The point is that you need to be able to balance the central politics of the party with local autonomy. This can be achieved by automating the compliance process and ensuring consistency between domain boundaries at the same time as the autonomous operation of the teams themselves. 7. The Self-Serve Data Platform And all of that happiness must be delivered on a self-service platform in order to allow teams to create and manage date products themselves. I recommend reading the whitepaper about the platform from CNCF. (1, 2 and 3)

Part 3: Infrastructure and Technical Architecture In this part, the author gives examples of building a self-service platform, as well as the design of the solution. 8. Comparing Self-Serve Data Platforms Google Cloud Platform, AWS, Databrics, self made over Kafka stack. The authors provide examples and compare the applicability of each. 9. Solution Architecture Design In this chapter, the authors explain how to work with functional and non-functional requirements, how to design a session, and how to use the C4 Model to model a system.

In general, the book seemed useful with a normal translation:)

#Data #Management #Processes #Architecture