Table of Contents

Search

  1. Abstract
  2. Supported Versions
  3. Tuning and Sizing Guidelines for Data Engineering Integration (10.4.x)

Tuning and Sizing Guidelines for Data Engineering Integration (10.4.x)

Tuning and Sizing Guidelines for Data Engineering Integration (10.4.x)

Case Study: Application Deployment

Case Study: Application Deployment

Application deployment requires communication between the Data Integration Service and the associated Model Repository Service. The Data Integration Service fetches the object and writes to the database schema of the Model Repository Service.

Environment

Chipset
Intel® Xeon® CPU E5-2680 v3 @ 2.50GHz
Cores
48 cores
Memory
132 GB
Operating system
Red Hat Enterprise Linux 7.6

Performance Chart

The following chart shows the time taken to deploy applications with a different number of objects:

Conclusions

Based on the case study, consider the following best practices for application deployment when you have a similar configuration:
Number of objects in applications
Having a large number of objects in the same application is not desirable. It increases the deployment time. It also increases the resource usage of the Data Integration Service and the Model Repository Service. Distributing objects in an optimal manner between various applications is key to achieve better performance.
Recommendation:
~ 50 objects per application.
Workaround for incremental application deployment
To deploy the changes made to the application, you stop the application and redeploy it. This process causes downtime. Applications must be designed to minimize the effects on this downtime. If the number of objects in the application are less, the effect of the downtime will be less severe.
Recommendation:
The number of objects in an application should not exceed ~ 60 objects per application.
Cursor requirements on the database
The process of application deployment needs to use cursors at the database layer (the schema associated with the Model Repository service). If applications are designed to be too large (1000+ of objects within) or with deep hierarchy of objects, the cursor usage will be greater at the database level.
Recommendation:
Required number of cursors >= Number of objects in application.

0 COMMENTS

We’d like to hear from you!