Databse architecture (single db vs client specific db) for Building Enterprise Web (RIA) application on cloud

by

The three main techniques that is usually applied to the database usage for this kind of a multi-tenant requirement is below. You have already specified some of them.

  1. Separate databases for each tenant:
    • very high cost, easy to maintain/customize, easy to tune, easy to backup, easy to code to.
  2. Shared database but different schema:
    • Low cost compared to (1), may encounter issues quickly with increased db size, easy to personalize per tenant, difficult to backup/restore per tenant, easy to code to.
  3. Shared database Shared schema:
    • Low cost, load of one tenant will affect others, security and app development a challenge, difficult to personalize per tenant, difficult to restore/backup.

I think the above points hold good for hosting on premise or on cloud.
If you see the number of tenants growing or the data getting bigger then 1) or 2) is better. I have used option 2) and have seen it helping development and maintenance.

More Related Contents:

  1. Microservices with shared database? using multiple ORM’s? [closed]
  2. Should each and every table have a primary key?
  3. Database development mistakes made by application developers [closed]
  4. Database, Table and Column Naming Conventions? [closed]
  5. What’s wrong with foreign keys?
  6. Surrogate vs. natural/business keys [closed]
  7. What is the best design for a database table that can be owned by two different resources, and therefore needs two different foreign keys? [closed]
  8. Normalization: What does “repeating groups” mean?
  9. Multiple schemas versus enormous tables [closed]
  10. First name, middle name, last name. Why not Full Name?
  11. Relational table naming convention [closed]
  12. What are the design criteria for primary keys?
  13. When I should use one to one relationship?
  14. Techniques for database inheritance?
  15. What is the recommended way to delete a large number of items from DynamoDB?
  16. Supertype-subtype database design
  17. In terms of databases, is “Normalize for correctness, denormalize for performance” a right mantra?
  18. How should international geographical addresses be stored in a relational database?
  19. Which is a good design for a database table that can be owned by two different resources, and therefore needs two different foreign keys? [closed]
  20. What’s wrong with nullable columns in composite primary keys?
  21. Best representation of an ordered list in a database?
  22. Ideas on database design for capturing audit trails [closed]
  23. NULL permitted in Primary Key – why and in which DBMS?
  24. Versioning Database Persisted Objects, How would you? [closed]
  25. Best design for a changelog / auditing database table? [closed]
  26. What are some best practices and “rules of thumb” for creating database indexes?
  27. What’s the better database design: more tables or more columns?
  28. Should OLAP databases be denormalized for read performance? [closed]
  29. Best beginner resources for understanding the EAV database model? [closed]
  30. Database – Data Versioning [closed]

Leave a Comment