AWS Certified Solutions Architect – ProfessionalDesign for New SolutionsHard
A software-as-a-service (SaaS) provider is building a new multi-tenant application that requires strict data isolation between tenants. Each tenant's data must be stored in a completely separate database instance to meet stringent compliance and security requirements. The application anticipates onboarding thousands of tenants, each requiring their own isolated environment. The solution must be cost-effective and scalable. Which multi-tenancy model and corresponding AWS database strategy should be chosen?
- ASiloed multi-tenancy model with Amazon Aurora Serverless V2 and a separate database instance per tenant.
- BPooled multi-tenancy model with Amazon Aurora Serverless V2 and a single database instance.
- CBridge multi-tenancy model with Amazon DynamoDB and separate tables per tenant.
- DSiloed multi-tenancy model with Amazon RDS for PostgreSQL and a separate database instance per tenant.
Show answer & explanationAnswer & explanation
Correct answer: A. Siloed multi-tenancy model with Amazon Aurora Serverless V2 and a separate database instance per tenant.
A siloed multi-tenancy model ensures strict data isolation by providing a separate database instance per tenant, fulfilling the compliance and security requirements. Aurora Serverless V2 is cost-effective and scalable for this model as it automatically scales capacity based on demand and charges per second, avoiding over-provisioning for thousands of individual instances.
Why the other options are wrong
- B. Pooled multi-tenancy shares a single database, which violates the strict data isolation requirement. Aurora Serverless V2 is good for cost-effectiveness but not with a shared instance in this scenario.
- C. Separate DynamoDB tables provide logical separation, but the question implies a relational database context ('database instance') and DynamoDB's data isolation is at the table level, not a full instance per tenant, which might not meet 'strict data isolation' for all compliance needs.
- D. While a siloed model with RDS for PostgreSQL meets isolation, managing thousands of separate RDS instances (even small ones) can be very expensive and operationally complex due to fixed instance costs and manual scaling, making it less cost-effective and scalable than Serverless V2.
Siloed Multi-Tenancy with Aurora Serverless V2
A SaaS architecture where each tenant has a dedicated database instance for strict data isolation, leveraging Aurora Serverless V2 for cost-effective, on-demand scalability.
- Each tenant gets a separate, isolated database instance.
- Ensures maximum data isolation for compliance and security.
- Aurora Serverless V2 provides automatic scaling and pay-per-second billing for cost-effectiveness.
- Ideal for applications with thousands of tenants requiring dedicated resources.
Memory trick: Siloed Isolation with Serverless Scales Smartly.