Building a Multi-Tenant Configuration System with Tagged Storage Patterns on AWS

Overview

Many contemporary SaaS applications operate in a multi-tenant environment, where each tenant shares a common infrastructure. Although this is an optimal design, it raises several issues regarding the configuration of individual applications. The multi-tenant architecture of AWS helps resolve this issue by implementing tagged storage.

In this strategy, the application dynamically selects different storage engines based on tags or prefixes defined by the developer. Unlike the previous strategy, which used only one storage engine, the tagged storage strategy leverages AWS services such as Amazon DynamoDB and Parameter Store to optimize performance based on the configuration type.

This blog provides a detailed explanation of the concept, architecture, and advantages of this design pattern.

Introduction

In a multi-tenancy architecture, an application provides services to many users while ensuring that their data remains separate. The architecture is very common in SaaS applications because it reduces costs and enables scalability.

Configuring such architectures can be quite challenging. Depending on what you want to do with the configuration, you may:

• Require fast access
• Require hierarchical configuration
• Require frequent updates

Conventional solutions would generally compel a developer to use:

• One storage service for all configurations (which will cause performance problems), or
• Several configuration services (and complicate things).

According to AWS, the solution to these problems is the concept of a tagged storage architecture.

Understanding the Tagged Storage Pattern

Tagged Storage Approach is one of the architectural methods used for assigning certain tags to configuration keys.

These tags help specify where to store and fetch configuration data.

Thus, through tagging, the request is routed automatically to the relevant storage system.

Key Components

1. Authentication and Request Processing

• The user is authenticated by using Amazon Cognito
• The request goes through the Amazon API Gateway and security layer

2. Microservices Level

• The configuration service operates as microservices (for example, on Amazon ECS/Amazon Fargate)
• The service processes the request and defines routing logic

3. Key-Prefixed Storage Routing

• Dependent on key prefixes, routing goes to:
  • Amazon DynamoDB for fast access
  • Parameter Store for structure configurations

4. Event-Driven Configuration Updates

• Takes advantage of EventBridge and Lambda
• Configurations are automatically updated without any interruptions

5. Caching Service

• Makes use of caching to improve application performance
• Events trigger cache updating

Key Design Concepts

Data Isolation for Multi-Tenancy

Each tenant’s data should be isolated. This will be done by using methods such as identifying each tenant (using JWT claims) so that there is never any data leakage between tenants.

Design Pattern for Dynamic Storage Selection

This application will use a strategy design pattern for storage selection. This means that storage will not be hardcoded but rather selected dynamically according to tags.

Event-Driven Approach to Handling Configuration Changes

The following are the benefits of using events for handling configuration updates:

• Changes trigger the events
• The cache and service are updated automatically

Thus, no restarts are necessary to effectuate changes.

Real-Time Configuration Updates Without Downtime

Using streaming techniques and event-driven configuration changes, zero downtime during configuration updates is ensured.

Benefits of Tagged Storage Pattern

Flexibility

Different types of configurations can use different storage systems based on their requirements.

Scalability

The system can efficiently handle multiple tenants and growing workloads.

Performance Optimization

High-frequency data is stored in fast-access systems, while structured data is stored in optimized services.

Cost Efficiency

By using the right storage for the right use case, unnecessary costs are avoided.

Operational Simplicity

Instead of managing multiple configuration services, a single system dynamically handles everything.

When to Use This Pattern

This architecture is useful when:

• You are building a multi-tenant SaaS application
• You have different types of configuration data
• You need real-time updates without downtime
• You want to optimize performance and cost simultaneously

Conclusion

It becomes apparent that the tagged storage pattern is an efficient way to handle configurations in multi-tenant systems. In essence, the use of dynamic routing to direct configuration information to the relevant storage medium enables the elimination of the limitations imposed by other configuration methods.

Drop a query if you have any questions regarding multi-tenant and we will get back to you quickly.

About CloudThat