The responsibility of the data architect

What does a data architect do? Just a design table? That's the first time I can think about it. But it turns out that it's not. The main responsibilities of the data architect are:
Data modeling and Analysis (45%)
Data loading and delivery (25%)
Data availability, performance, security assurance (15%)
Data quality and governance (10%)
Other (5%)

1. Data Modeling and analysis

Understanding the data (understand)
Understanding data in an organization is the most important responsibility. If he works in a securities agency, he needs to know what is the professional term for profitability. An architect who doesn't really understand the meaning of data cannot create a fully functional function.

Data Modeling (modelling)
It does not just design database tables. It also contains XML files and schemas (XSD), OLAP cubes, interface file format definitions, and so on. Not only is the table design also includes indexes, logs, partitions, databases and data servers, triggers, audits, and so on. Physical data model and logical data model. Although we often ignore the physical composition of the database, we hand this part of the work to the DBA.

Data analysis

We often hand over data analysis to a dedicated data analyst. We always think that our job is to design data model, business analysis is not my job content. This is a wrong point of view. As a data architect, you need to know everything. can help us better design the data architecture.

2. Data loading

Data is not able to run to our database in accordance with our requirements, but requires a complex process to update/INSERT into our Data Warehouse step-by-step. We need to understand the detailed details of the entire data load. Understand the entire loader architecture, and what they do every step of the process. If you use a scheduling tool such as Autosys, understand how it works. How to create a conditional alert, a loop, how to detect load time, and so on. Of course, many of these tasks are done by the Product Support Team (Production), but designing the entire loaded process/framework is the responsibility of the data architect.

If a data architect knows only the data model and does not understand the process of loading the data, then he is no different from the blind. When and how to load data into a database, there will be significant changes to those data.

*************************************

(not to be continued) ....

The responsibility of the data architect