1.System architecture before migration.
The system includes:
- 3 Heroku dynos (2 dynos running nodejs and 1 dyno running java)
- 1 Heroku postgres (Database)
- 1 Heroku private space
2.Reason for migration
- The old system is connecting to S3 but the system is located on Heroku, which is difficult to centrally manage
- The price of Heroku private space is high from 1000$ - 5000$ which is not suitable for websites with only about 200 to 300 people accessing at the same time.
After successful conversion, the cost is reduced to 1/3 compared to Heroku
3.Architecture on AWS
The system includes:
- 03 App runner (2 app runner running nodejs and 1 app runner running java)
- Amazon RDS for PostgreSQL
- AWS WAF
- Interface vpc endpoint and Gateway vpc endpoint
4.Notes when using App runner
4.1. Use github to configure CI/CD with App runner
Refer to the article
4.2. Use app runner vpc connector
Refer to the article
4.3. Use interface vpc endpoint and gateway vpc endpoint for
S3
Refer to the article
4.4. Use WAF to allow access to only certain IP addresses
access the system
Refer to the article
4.5. Use the DBeaver tool to easily migrate data from Heroku
to RDS
Refer to the article
5.Deep Dive on AWS App Runner VPC Networking
Refer to the article
If you have any questions, please contact us: https://www.linkedin.com/in/huydanggdg/
Top comments (0)