High-Level Design (HLD) System Design Cheatsheet

Ankush kunwar

2 min readFeb 1, 2025

1. Understanding Requirements

Functional Requirements: Core features (e.g., user login, messaging, video streaming, etc.).
Non-Functional Requirements: Scalability, availability, fault tolerance, consistency, etc.
Constraints & Assumptions: Users, data size, traffic volume, latency requirements.

2. Defining System Components

Client Layer: Web, mobile apps, APIs.
Application Layer: Microservices, monolith, API gateways.
Data Layer: Databases (SQL/NoSQL), caching (Redis, Memcached).
Storage Layer: Blob storage, file storage, CDN.
Networking: Load balancers, reverse proxies.

3. Database Selection

Relational Databases (SQL): ACID compliance, structured data.
NoSQL Databases: Document-based (MongoDB), Key-value (Redis), Columnar (Cassandra), Graph (Neo4j).
Sharding & Partitioning: Horizontal scaling.
Replication: Master-slave, leader-follower.
Indexing & Query Optimisation.

4. Scalability & Performance

Load Balancing: Round-robin, Least connections, Consistent hashing.

Caching Strategies:

CDN for static content.
Application caching (Redis, Memcached).
Database caching (Read replicas, materialized views).

Asynchronous Processing:

Message queues (Kafka, RabbitMQ, SQS).
Event-driven architecture.

5. Availability & Fault Tolerance

Replication: Multi-region, Multi-master, Read replicas.
Failover & Redundancy: Hot standby, cold standby.
Distributed Systems Principles: CAP Theorem (Consistency, Availability, Partition Tolerance).

6. Consistency & Data Integrity

Strong vs Eventual Consistency.
Distributed Transactions: Two-phase commit, Saga pattern.
Quorum-based Voting (Raft, Paxos).

7. Security Considerations

Authentication & Authorization: OAuth, JWT, OpenID.
Data Encryption: TLS, SSL, AES.
Rate Limiting & API Security.
DDOS Protection, Firewalls, IDS/IPS.

8. Monitoring & Logging

Metrics Collection: Prometheus, Grafana.
Logging & Analysis: ELK Stack (Elasticsearch, Logstash, Kibana), Splunk.
Alerting & Incident Management: PagerDuty, Opsgenie.

9. Common System Design Problems

URL Shortener: Hashing, unique ID generation.
Rate Limiting: Token bucket, leaky bucket.
Messaging Queue: Kafka, RabbitMQ, SQS.
Designing Instagram/Twitter: Timeline generation, fan-out.
Designing Netflix/YouTube: Content delivery, CDN.
Designing Uber: Real-time tracking, geospatial indexing.

10. Interview Tips

Clarify Requirements First.
Start with a High-Level Architecture before diving deep.
Discuss Trade-offs (CAP theorem, consistency vs availability).
Think About Scaling Early.
Use a Whiteboard or Paper to Diagram Your Solution.
Keep It Conversational & Collaborative.

Final Note:

System design interviews are about demonstrating structured thinking, scalability trade-offs, and handling real-world constraints effectively. Practice designing different systems and focus on reasoning over perfect solutions!

Thank you for reading !!!

If you enjoy this article and would like to Buy Me a Coffee, please click here.

you can connect with me on Linkedin.