Pull vs Push CDN Models
What you'll learn: How CDNs populate content on edge servers using either on-demand fetching (pull) or pre-populated distribution (push).
CDNs need to get content from your origin server to their edge locations. There are two fundamental approaches: pull (lazy loading) and push (pre-population).
Pull CDN Model
In a pull model, edge servers cache content on-demand. When a user requests a file that isn't cached yet (a cache miss), the edge server fetches it from the origin, caches it locally, then serves it to the user. Subsequent requests for that same content hit the cache.
Analogy: Think of a library branch that orders books from a central warehouse only when someone requests them. The first person waits longer, but everyone after gets instant access.
Characteristics:
- Origin remains authoritative source
- Minimal upfront work required
- Edge servers populate gradually based on real user traffic
- First request per edge location suffers higher latency
- Works well with unpredictable or large content catalogs
Push CDN Model
In a push model, you proactively upload content to edge servers before users request it. You control when and what gets distributed across the CDN network.
Analogy: A library branch that stocks popular books in advance, ensuring they're instantly available when anyone walks in.
Characteristics:
- You explicitly push content to the CDN
- Faster initial response times (no cold-start penalty)
- Requires predicting what content to distribute
- Better control over cache contents
- Ideal for known, high-value content (software releases, scheduled videos)
Which to Choose?
Pull works for dynamic sites with unpredictable traffic patterns. Push excels when you know exactly what content will be popular and when (product launches, live event recordings).
Key Takeaway: Pull CDNs cache content lazily on first request, while push CDNs require you to pre-populate edge servers—choose based on traffic predictability and control needs.