What are CDN caching strategies? How to optimize CDN cache hit rate? - 面试题

Core Concepts of CDN Caching Strategies

CDN caching strategies determine how long content is stored on edge nodes and how it's updated, directly affecting user experience and origin server load. Proper caching strategies can maximize the performance benefits of CDN.

Main Caching Strategies

1. TTL-Based Caching (Time To Live)

TTL is the most basic cache control mechanism, specifying how long content stays cached on CDN nodes:

Short TTL (seconds): For frequently updated content like news, stock data
Medium TTL (minutes): For occasionally updated content like product info, user profiles
Long TTL (hours/days): For rarely changing content like static assets, images, videos

Setting methods:

http
Cache-Control: max-age=3600  // Cache for 1 hour
Cache-Control: max-age=86400 // Cache for 1 day

2. Cache Key Configuration

Cache keys determine which requests are treated as the same content:

Default cache key: Based on full URL
Custom cache key: Can include or exclude specific request headers, query parameters
Ignore query parameters: For parameters that don't affect content (like ?timestamp=xxx)

Example:

Full URL: https://example.com/image.jpg?width=800&quality=90
Ignore quality: https://example.com/image.jpg?width=800

3. Hierarchical Caching Strategy

CDN typically uses multi-level caching architecture:

Edge node cache: Closest to users, smaller capacity, fastest response
Regional node cache: Covers specific regions, medium capacity
Origin cache: Largest capacity, final backup

Cache lookup order: Edge node → Regional node → Origin

4. Cache Warming

Pre-populate CDN nodes with content before official release:

Active warming: Push content via API or management console
Passive warming: Trigger caching through simulated user requests
Use cases: Major events, new releases, popular content

5. Cache Purging

Actively clear cached content from CDN nodes:

URL purge: Clear cache for specific URL
Directory purge: Clear cache for entire directory
Full site purge: Clear all caches (use with caution)

Purge methods:

Immediate purge: Clear cache immediately
Soft purge: Don't renew after TTL expires

Advanced Caching Strategies

1. Dynamic Content Caching

Even dynamic content can be cached through:

Negotiation caching: Using ETag or Last-Modified headers
Edge computing: Execute simple dynamic logic at CDN edge nodes
API response caching: Cache API response results

2. Intelligent Caching

Automatically adjust caching strategies based on content characteristics:

Auto long cache for static files: Automatically set long TTL based on file extension
Auto short cache for dynamic content: Automatically shorten TTL when frequent updates detected
Priority caching for popular content: Optimize cache based on access frequency

3. Conditional Caching

Decide whether to cache based on specific conditions:

Based on user type: Cache for regular users, real-time for paid users
Based on geography: Cache in some regions, real-time in others
Based on time: Cache during peak hours, real-time during off-peak

Caching Strategy Optimization Recommendations

1. Set TTL Reasonably

Static resources: Set long TTL (1 day to 1 year)
Dynamic content: Set short TTL (seconds to minutes)
Versioned resources: Use filename versioning, can set infinite cache

2. Use Cache Control Headers

http
// Force caching
Cache-Control: public, max-age=31536000, immutable

// Prohibit caching
Cache-Control: no-store, no-cache

// Negotiation caching
Cache-Control: no-cache
ETag: "abc123"
Last-Modified: Wed, 21 Oct 2026 07:28:00 GMT

3. Monitor Cache Hit Rate

High hit rate (>90%): Caching strategy is effective
Medium hit rate (70-90%): Needs optimization
Low hit rate (<70%): Need to reevaluate strategy

4. A/B Test Caching Strategies

Test different caching strategies on different user groups:

Test different TTL values
Test different cache key configurations
Test whether to enable edge computing

Common Issues and Solutions

Issue 1: Users still see old content after update

Solutions:

Use URL versioning (like style.v2.css)
Actively purge CDN cache
Set reasonable TTL

Issue 2: Low cache hit rate

Solutions:

Check if cache key configuration is reasonable
Analyze reasons for non-caching (like Cookie, Authorization headers)
Adjust TTL settings

Issue 3: Origin server still under high pressure

Solutions:

Increase TTL to extend cache time
Enable cache warming
Check if there are many origin requests

Interview Points

When answering this question, emphasize:

Understanding of different caching strategies and their use cases
Ability to choose appropriate strategies based on business needs
Understanding of how caching strategies affect performance
Practical optimization experience or case studies
Ability to analyze cache hit rates and suggest improvements