Software Architecture

Step 2: Assign Tables to Data Domains

The next step is to group tables along a specific bounded context, assigning tables that belong to a specific data domain into their own schema. A schema is a logical construct in database servers. A schema contain objects such as tables, views, functions, and so on. In some database servers, like Oracle, the schema is same as the user, while in other databases, like SQL Server, a schema is logical space for database objects where users have access to these schemas.
As illustrated in Figure 6-21, we have created schemas for each data domain and moved tables to the schemas to which they belong.

Figure 6-21. Services use the primary schema according to their data domain needs

When tables belonging to different data domains are tightly coupled and related to one another, data domains must necessarily be combined, creating a broader bounded context where multiple services own a specific data domain. Combining data domains is discussed in more detail in Chapter 9.

Data Domain Versus Database Schema

A data domain is an architectural concept, whereas a schema is a database construct that holds the database objects belonging to a particular data domain. While the relationship between a data domain and a schema is usually one to one, data domains can be mapped to one or more schemas, particularly when combining data domains because of tightly coupled data relationships. We will be referring to a data domain and schema to mean the same thing, and will be using the terms interchangeably.
To illustrate the assignment of tables to schemas, consider the Sysops Squad example where the billing table must be moved from its original schema to another data domain schema called payment:
ALTER

SCHEMA

payment

TRANSFER

sysops
.
billing
;

Alternatively, a database team can create synonyms for tables that do not belong in their schema. Synonyms are database constructs, similar to symlink, that provide an alternate name for another database object that can exist in the same or different schema or server. While the idea of synonyms is to eliminate cross-schema queries, read or write privileges are needed to access them.

To illustrate this practice, consider the following cross-domain query:
SELECT

history
.

ticket_id
,

history
.

notes
,

agent
.

name