How do you decide what data should be cached and for how long?

Question

Anubhav Kumar · Answer

How to Decide What Data to Cache and for How Long?
Caching is a trade-off between freshness and performance. To decide what data to cache and for how long, consider the following factors:
1. What Data Should Be Cached?
Good Candidates for Caching
Data Type
Reason for Caching
Frequently accessed data
Avoid redundant database or API calls (e.g., top articles, user profile).
Expensive computations
Save CPU time (e.g., complex reports, aggregated statistics).
Static or rarely changing data
No need to fetch frequently (e.g., product categories, country lists).
Third-party API responses
Reduce external API calls and costs (e.g., weather, currency exchange rates).
Session-specific data
Improve user experience (e.g., shopping cart, user preferences).
Bad Candidates for Caching
Data Type
Why Not Cache?
Rapidly changing data
Cached values might become stale (e.g., stock prices, live scores).
Sensitive or personal data
Security and privacy risks.
Data with strict consistency needs
Can lead to stale reads (e.g., banking transactions).
and 
2. How Long Should Data Be Cached?
Cache duration depends on how frequently the data changes and how critical freshness is.
Data Type
Suggested Cache Duration
Static content (logos, CSS, JS)
Days to weeks (max-age=604800)
Product lists, categories
Hours to a day (max-age=3600 to 86400)
User sessions, shopping carts
Session duration (Sliding Expiration)
News articles, blog posts
Minutes to hours (max-age=600 to 3600)
Real-time stock prices, chat messages
Do not cache (or use very short TTL andlt; 10s)
Caching Strategies
Strategy
When to Use?
Absolute Expiration (max-age)
Good for data that updates at known intervals (e.g., daily reports).
Sliding Expiration
Ideal for session-related data (e.g., extend expiry as long as the user is active).
ETag / Last-Modified
Best for browser caching (ensure the latest data is fetched only if changed).
Cache Invalidation
Use when data needs immediate updates (e.g., clear cache when an order is placed).
and 
3. Choosing Cache Location
Cache Type
Use Case
MemoryCache
Fastest, best for single-server apps.
Distributed Cache (Redis, SQL Server Cache)
Best for multi-instance apps.
Browser Cache (Cache-Control, ETag)
Best for static assets and API responses.
CDN (Content Delivery Network)
Best for images, scripts, videos.
and 
Summary
Cache frequently accessed, expensive, or static data.Avoid caching real-time, sensitive, or fast-changing data.Set cache duration based on data volatility (seconds for real-time, hours/days for static data).Use a combination of in-memory, distributed, and browser caching for efficiency.Implement cache invalidation strategies for dynamic content.
and

Data Type	Suggested Cache Duration
Static content (logos, CSS, JS)	Days to weeks (`max-age=604800`)
Product lists, categories	Hours to a day (`max-age=3600 to 86400`)
User sessions, shopping carts	Session duration (`Sliding Expiration`)
News articles, blog posts	Minutes to hours (`max-age=600 to 3600`)
Real-time stock prices, chat messages	Do not cache (or use very short TTL < 10s)

interview

How do you decide what data should be cached and for how long?

Anubhav Kumar

Can you answer this question?

1 Answers

How to Decide What Data to Cache and for How Long?

1. What Data Should Be Cached?

2. How Long Should Data Be Cached?

3. Choosing Cache Location

Summary

Liked By

Data Type	Reason for Caching
Frequently accessed data	Avoid redundant database or API calls (e.g., top articles, user profile).
Expensive computations	Save CPU time (e.g., complex reports, aggregated statistics).
Static or rarely changing data	No need to fetch frequently (e.g., product categories, country lists).
Third-party API responses	Reduce external API calls and costs (e.g., weather, currency exchange rates).
Session-specific data	Improve user experience (e.g., shopping cart, user preferences).

Data Type	Why Not Cache?
Rapidly changing data	Cached values might become stale (e.g., stock prices, live scores).
Sensitive or personal data	Security and privacy risks.
Data with strict consistency needs	Can lead to stale reads (e.g., banking transactions).

Strategy	When to Use?
Absolute Expiration (`max-age`)	Good for data that updates at known intervals (e.g., daily reports).
Sliding Expiration	Ideal for session-related data (e.g., extend expiry as long as the user is active).
ETag / Last-Modified	Best for browser caching (ensure the latest data is fetched only if changed).
Cache Invalidation	Use when data needs immediate updates (e.g., clear cache when an order is placed).

Cache Type	Use Case
MemoryCache	Fastest, best for single-server apps.
Distributed Cache (Redis, SQL Server Cache)	Best for multi-instance apps.
Browser Cache (Cache-Control, ETag)	Best for static assets and API responses.
CDN (Content Delivery Network)	Best for images, scripts, videos.