Design DynamoDB-like Store

Question

Design the architecture for DynamoDB-like Store. Cover scalability, data storage, caching, and real-time requirements.

Codemia · Accepted Answer

### Functional Requirements
1. **Data Storage**: The store should support key-value pair storage with flexible schema.
2. **CRUD Operations**: It must support Create, Read, Update, and Delete operations on items.
3. **Query Support**: The ability to query by primary key and secondary indexes.
4. **Global Distribution**: Support for multi-region deployments, allowing low-latency access globally.
5. **Data Consistency**: Options for strong and eventual consistency.
6. **Rate Limiting**: Implement API rate limiting to prevent abuse.

### Non-Functional Requirements
1. **Scalability**: Must handle millions of requests per second and scale horizontally.
2. **Availability**: 99.99% uptime SLA, ensuring the system is resilient to failures.
3. **Latency**: Read and write latency should be less than 10 ms.
4. **Security**: Data encryption at rest and in transit, plus IAM-based access control.

Assuming a target of 1 million users with an average of 100 read and 10 write requests per user per day:

- **Total Requests per Day**: 
  - Reads: 1,000,000 users * 100 reads = 100,000,000 reads/day
  - Writes: 1,000,000 users * 10 writes = 10,000,000 writes/day

- **Total Requests per Second**: 
  - Reads: 100,000,000 / 86400 seconds = ~1,157 reads/sec
  - Writes: 10,000,000 / 86400 seconds = ~116 writes/sec

- **Storage Estimation**: Assuming an average item size of 1 KB:
  - Total Data: (1,000,000 users * 10 items/user) * 1 KB/item = ~10 GB

- **Scaling Factor**: Plan for a 3x growth over 3 years, leading to ~30 GB of storage and ~3,471 reads/sec and ~348 writes/sec.

### Component Diagram
- **API Gateway**: Amazon API Gateway for routing requests, rate limiting, and security.
- **Load Balancer**: Elastic Load Balancer (ELB) to distribute incoming traffic across multiple instances.
- **Application Servers**: AWS Lambda or EC2 instances running the application logic.
- **DynamoDB**: Core data storage, leveraging its global tables for multi-region support.
- **Caching Layer**: Amazon ElastiCache (Redis) for caching frequently accessed data.
- **Monitoring**: Amazon CloudWatch for logging and metrics, with Alarms for critical thresholds.

### Technology Choices
- **Language**: Node.js or Python for serverless functions.
- **Data Format**: Use JSON for data interchange, considering its natural fit with NoSQL.

### Schema Design
- **Table: Users**
  - `UserID` (Primary Key, String)
  - `Name` (String)
  - `Email` (String, Unique)
  - `CreatedAt` (Timestamp)

- **Table: Items**
  - `ItemID` (Primary Key, String)
  - `UserID` (Secondary Index, String)
  - `Data` (Map)
  - `Timestamp` (Timestamp)

- **Access Patterns**
  - Retrieve user by `UserID`.
  - Query items by `UserID` to fetch user's items.
  - Use global secondary index on `CreatedAt` for time-based queries.

### Architectural Decisions
1. **CAP Theorem**: Opt for eventual consistency for high availability and partition tolerance, allowing for faster writes and reads.
2. **Global Distribution vs. Latency**: Trade-off between global distribution (using DynamoDB global tables) and potential latency issues during eventual consistency.
3. **Caching Strategy**: Using ElastiCache introduces complexity in cache invalidation but can drastically improve read performance.
4. **Rate Limiting**: Implementing rate limiting introduces a bottleneck but prevents abuse and ensures fair usage across users.
5. **Deployment Strategy**: Rolling updates will mitigate downtime during deployments but require careful versioning management to ensure backward compatibility.

Design DynamoDB-like Store

Amazon

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

Functional Requirements

Capacity Estimation

Submit Your Answer

Amazon Product Manager 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