Caching Design

[TOC]

Eviction Strategies

Least Recently Used(LRU)

Advantages:

Easy Implementation
Efficient Use of Cache
Adaptability

Disadvantages:

Strict Ordering
Cold Start Issues
Memory Overhead

Least Frequently Used(LFU)

Advantages:

Adaptability to Varied Access Patterns
Optimized for Long-Term Trends
Low Memory Overhead

Disadvantages:

Sensitivity to Initial Access
Difficulty in Handling Changing Access Patterns
Complexity of Frequency Counters

First-In-First-Out(FIFO)

Advantages:

Simple Implementation
Predictable Behavior
Memory Efficiency

Disadvantages:

Lack of Adaptability
Inefficiency in Handling Variable Importance
Cold Start Issues

Random Replacement

Advantages:

Simplicity
Avoids Biases
Low Overhead

Disadvantages:

Suboptimal Performance
No Adaptability
Possibility of Poor Hit Rates

Caching For API

Caching APIs can significantly improve performance in system design by addressing several key factors:

Faster Data Retrieval
Reduced Database Load
Minimized Network Latency
Enhanced Throughput
Improved User Experience
Resource Optimization
Decreased API Rate Limiting
Scalability

Caching APIs reduces server load in system design through several mechanisms:

Serving Repeat Requests from Cache
Decreasing Database Queries
Reducing Computational Work
Handling Spikes in Traffic
Efficient Use of Resources
Enhanced System Stability and Reliability

Client-Side Caching

Benefits:

Reduces server load by storing responses directly on the client.
Decreases latency since the data is fetched from the client's local storage.

Use Cases:

Static assets like images, CSS, and JavaScript files.
API responses that change infrequently, such as user profile data.

Server-Side Caching

Benefits:

Reduces the need to recompute responses for repeated requests.
Can handle a large number of requests efficiently.

Use Cases:

Frequently accessed data like product catalogs or new feeds.
API responses that are resource-intensive to generate.

Reverse Proxy Caching

Benefits:

Caches responses at the network edge, reducing latency and load on the origin server.
Improves response times for end-users.

Use Cases:

Publicly accessible APIs with high traffic volumes.
Content delivery networks(CDNs) for static and dynamic content.

Distributed Caching

Benefits:

Spreads the cache across multiple nodes, improving scalability and fault tolerance.
Maintains data availability in the event that a node fails.

Use Cases:

Large-scale applications with significant amounts of data to cache.
Systems requiring high availability and reliability.

Application-Level Caching

Benefits:

Customizable caching strategies based on application logic.
Can be integrated directly into the application code.

Use Cases:

Specific parts of an application that require fine-grained control over caching.
Scenarios where data validity and freshness need to be closely managed.

Database Caching

Benefits:

Offloads database queries, improving database performance.
Can cache query results or specific database rows.

Use Cases:

Frequently queried database tables.
Complex queries that require significant computation.

Caching Strategy

Cache Aside

TODO

Read Through

TODO

Write Around

TODO

Write Back

TODO

Write Through

TODO

Challenges

Cache Miss Attack

Cache Miss Attack: It refers to the scenario where data to fetch doesn't exist in the database and the data isn’t cached either. So every request hits the database eventually, defeating the purpose of using a cache. If a malicious user initiates lots of queries with such keys, the database can easily be overloaded.