Design Email Infrastructure for mobile apps

Question

Design a event-driven email system that handles millions of requests. Discuss trade-offs in consistency, availability, and performance.

Codemia · Accepted Answer

- **Functional Requirements**:
  - Support sending and receiving emails through mobile applications.
  - Allow users to schedule emails for future sending.
  - Provide real-time notifications for incoming emails.
  - Support for rich media attachments (e.g., images, documents).
  - Implement a search feature for retrieving emails.

- **Non-Functional Requirements**:
  - **Scalability**: Must handle 10 million email requests per day with a peak load of 1000 requests per second.
  - **Availability**: Aim for 99.99% uptime with multi-region redundancy.
  - **Performance**: Email delivery should occur within 1-2 seconds.
  - **Security**: Implement OAuth 2.0 for authentication and TLS for email transmission.
  - **Data Consistency**: Support eventual consistency for email status updates.

To estimate capacity:
- **Daily Email Volume**: 10 million emails/day.
- **Peak Load**: 1000 requests/second.

- **Storage Estimation**: Assuming an average email size of 75 KB (including headers and attachments):
  - Daily storage = 10 million * 75 KB = 750 GB/day.
  - Monthly storage = 750 GB/day * 30 days = 22.5 TB.

- **Database Operations**: Each email operation (send, receive, status update) can be estimated at around 5 operations. Therefore:
  - Total operations per day = 10 million emails * 5 = 50 million operations.
  - Average operations per second = 50 million / 86400 = ~580 operations/second.

- **Cache Size**: Assuming 10% of emails are frequently accessed, we need to cache 1 million emails, which could be around 75 GB.

The architecture consists of the following components:
- **API Gateway**: AWS API Gateway for managing and routing requests.
- **Microservices**:
  - **Email Service**: Responsible for sending and receiving emails (Node.js).
  - **Notification Service**: Real-time push notifications to mobile apps (Firebase Cloud Messaging).
  - **Scheduler Service**: Handles scheduling of emails (using Quartz Scheduler).
- **Database**: Amazon DynamoDB for storing email metadata and user data, chosen for its scalability and performance.
- **Cache Layer**: Redis for caching frequent email data and user sessions.
- **Message Queue**: Amazon SQS for decoupling services and handling email sending asynchronously.
- **Load Balancer**: AWS Elastic Load Balancer to distribute incoming traffic across instances.

The data model includes the following tables in DynamoDB:
- **Users**:
  - `userId` (PK)
  - `email` (String)
  - `oauthToken` (String)
  - `createdAt` (Timestamp)

- **Emails**:
  - `emailId` (PK)
  - `senderId` (FK to Users)
  - `recipientId` (FK to Users)
  - `subject` (String)
  - `body` (String)
  - `attachmentUrls` (List of Strings)
  - `status` (String: SENT, RECEIVED, SCHEDULED)
  - `createdAt` (Timestamp)
  - `scheduledSendAt` (Timestamp)

- **EmailMetadata** (for caching):
  - `emailId` (PK)
  - `lastAccessed` (Timestamp)
  - `accessCount` (Number)

This schema supports efficient querying for user emails and facilitates real-time updates.

- **Consistency vs. Availability**: Choosing eventual consistency for email status updates allows for higher availability and better performance, as emails can be processed quickly without waiting for all nodes to be updated.

- **Database Choice**: Opting for DynamoDB provides scalability but may introduce complexity in querying. For operations that require complex joins, we may need to denormalize data.

- **Caching Strategy**: Using Redis as a cache improves performance for frequently accessed data but introduces the challenge of cache invalidation, especially with email status changes.

- **Global Deployment**: Deploying services in multiple regions improves latency but introduces challenges in data replication and conflict resolution, requiring careful design of the consistency model to handle stale data.

Design Email Infrastructure for mobile apps

Oracle

What the Interviewer Expects

Key Topics to Cover

How to Approach This

Possible Follow-up Questions

Practice a Similar Problem on Codemia

Sample Answer

Requirements

Capacity Estimation

Submit Your Answer

Oracle Software Engineer Interview Guide

Related Questions

Design a high-throughput Inventory Management System

Design a low-latency Rate Limiting System

Design a fault-tolerant Payment System

Design a fault-tolerant Messaging System

Design Walmart Product Search