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

Question

Anubhav Sharma · 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). 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 &lt; 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). 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. 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.

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)

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.

interview

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

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