Identify Application Workload
On this page
The first step in the schema design process is to identify the operations that your application runs most frequently. Knowing your application's most common queries helps you create effective indexes and minimize the number of calls the application makes to the database.
About this Task
When you consider your application's workload, consider the scenarios your application currently supports and scenarios it may support in the future. Design your schema to function in all stages of your application development.
Steps
Identify the data your application needs
To identify the data that your application needs, consider the following factors:
Your application's users and the information they need.
Your business domain.
Application logs and frequently-run queries. To see database commands run on a MongoDB deployment, see Database Profiler.
Create a workload table with your application's queries
Fill out the following table with the queries that your application needs to run:
Action | Query Type | Information | Frequency | Priority |
|---|---|---|---|---|
The action that a user takes to trigger the query. | The type of query (read or write). | The document fields that are either written or returned by the query. | How frequently your application runs the query. Queries that are run frequently benefit from indexes and should be optimized to avoid lookup operations. | How critical the query is to your application. |
Example
The following example shows a workload table for a blog application:
Action | Type | Information | Frequency | Priority |
|---|---|---|---|---|
Submit a new article | Write | author, text | 10 per day | High |
Submit a comment on an article | Write | user, text | 1,000 per day (100 per article) | Medium |
View an article | Read | article id, text, comments | 1,000,000 per day | High |
View article analytics | Read | article id, comments, clicks | 10 per hour | Low |
Next Steps
After you identify your application's workload, the next step in the schema design process is to map related data in your schema. See Map Schema Relationships.