CDN Explained: How Content Delivery Networks Work

Picture this. A company runs its only server in New York, and someone in India opens their website:

• The page sits on that one server, all the way across the world.
• So every image, every file, has to travel halfway around the planet and back.
• The user just stares at a loading spinner for ages, right?

Now what if there was a copy of that website sitting much closer to the user, like in their own city? It would load almost instantly. That’s the whole idea behind a CDN, and by the end of this lesson you’ll know exactly how it pulls that off.

🎯 The Problem

Let’s start with the pain a CDN is built to solve. Say Alex runs a website with just one server somewhere far away. Here’s what goes wrong:

• That one server is called the origin server. The origin server is the original home of your content, the place where the real files actually live.
• If it sits in New York, then a user in India is thousands of kilometers away. Data takes time to travel that distance, and that travel time is called latency. Latency is just the delay between asking for something and getting it back.
• The farther the server, the higher the latency, so the site feels slow no matter how good your code is.

And distance isn’t even the only problem. There’s the load issue too:

• Every single user in the world hits that same one server. So when traffic spikes, the server gets overwhelmed by requests.
• A slow or overloaded origin means slow pages for everyone, and if it crashes, the whole site goes down.

So we’ve got two problems sitting on one machine: it’s too far from most users, and it has to handle everyone alone. A CDN fixes both at once.

🏪 Real-World Analogy

Think about a big shop, like a warehouse store. Imagine there’s only one giant warehouse for the entire country:

• Every time you want to buy milk, you’d have to drive all the way to that one warehouse, maybe in another state.
• That’s a long trip for something simple, right? And imagine everyone in the country driving to the same place. Total chaos.

Now think about how it actually works in real life:

• Instead of one faraway warehouse, there are small convenience store branches near you, like one in every neighborhood.
• The stuff people buy most often, milk, bread, snacks, is already stocked on those local shelves.
• So you just walk to the corner shop and grab it. Fast and easy.

A CDN works exactly like those local branches. The faraway warehouse is the origin server, and the neighborhood shops are servers placed close to users, each keeping copies of the popular stuff. Keep this picture in your head, every part below maps back to it.

🌍 What is a CDN?

So let’s define it properly now. A CDN, short for Content Delivery Network, is a network of servers spread out across the world that keep copies of your content close to your users. Here are the two pieces you need to know:

• Origin server. This is the original source, the one true home of your files. When the CDN doesn’t have something, this is where it goes to get it.
• Edge server. These are the CDN’s servers placed all around the globe, near where users actually are. An edge server is called “edge” because it sits at the edge of the network, close to the user, not deep in some central data center.

And here’s the key idea that ties it together:

• The edge servers keep cached copies of your content. Caching just means saving a copy somewhere closer or faster so you don’t have to fetch the original every time.
• So a user in India gets served from a nearby edge server in India, not from the origin all the way in New York.
• That short trip is what makes the site feel fast.

The whole point is distance. Less distance means less latency, and less latency means a faster site.

⚙️ How a CDN Works

Okay so how does a request actually flow through a CDN? Let’s walk it step by step. When a user asks for a file, like an image, here’s what happens:

• The request doesn’t go to the faraway origin. Instead it goes to the nearest edge server, the one closest to the user.
• That edge server checks: “Do I already have a copy of this file saved?”
• If yes, that’s called a cache hit. The edge server just sends back its saved copy right away. Super fast, because the origin isn’t even touched.
• If no, that’s a cache miss. The edge server doesn’t have it, so it fetches the file from the origin server once, sends it to the user, and saves a copy for next time.

So the first user to ask for something might wait a little, because it’s a miss and the edge has to go get it. But every user after that gets a fast hit. Here’s the flow as a picture:

And here’s the difference the CDN makes, the near path versus the far path:

📦 What CDNs Cache

Now here’s an important thing. A CDN is great for some kinds of content and not so great for others. The split is between static and dynamic content:

• Static content is stuff that’s the same for everyone and doesn’t change often. Think images, CSS files, JavaScript files, videos. The same logo gets sent to every user, so caching it once at the edge is perfect.
• Dynamic content is stuff that’s different for each user or changes constantly. Think your personal account page, your bank balance, a live cricket score. You can’t just cache one copy, because everyone needs a different answer.

So as a rule, CDNs love static content and struggle with highly personalized or live data. Here’s the breakdown:

You see this everywhere in real apps:

• Netflix serves its video streams from edge servers close to you, so the movie plays smoothly without buffering.
• YouTube serves its thumbnails and video files from a CDN, because millions of people load the same ones.
• Pretty much every website serves its images and stylesheets through a CDN, since those files don’t change per user.

⚡ Benefits

So why do almost all big sites use a CDN? Let’s lay out what you actually get:

• Lower latency. Content comes from a server near the user instead of one across the world, so pages and videos load faster.
• Less load on the origin. Most requests get answered at the edge, so your origin server handles far fewer hits. It can breathe.
• Handles traffic spikes. When a video goes viral, the many edge servers soak up the flood together. One origin alone could never handle that crowd.
• Some DDoS protection. A DDoS attack is when attackers flood a site with fake traffic to knock it offline. Because a CDN has lots of big edge servers, it can absorb and spread out a lot of that junk traffic before it ever reaches your origin.
• Saves bandwidth costs. Serving from the edge often costs less than serving everything straight from your origin.

The short version: faster for users, easier on your servers, and more resilient when things get crazy.

⚠️ Things to Watch

A CDN isn’t magic though. There are real things to be careful about, and most of them come down to one word: freshness.

• Stale content. Because the edge keeps a saved copy, it can serve an old version after you’ve updated the original. Say Alex fixes a typo in the logo, but the edge still hands out the old logo until its copy expires. That old saved copy is called stale content.
• Cache invalidation. To fix staleness, you tell the CDN “this file changed, drop your old copy.” Forcing the edge to throw away its cached copy is called a cache purge or invalidation. After a purge, the next request becomes a miss and fetches the fresh version.
• A common trick is to rename files when they change, like style.v2.css instead of style.css. A new name looks like a brand new file to the CDN, so it fetches the fresh one automatically.
• Not for personalized data. Don’t try to cache things that are unique per user or change every second. Your bank balance behind a CDN would be a disaster, right? Keep dynamic, personal stuff going to the origin.

⚠️ Common Mistakes and Misconceptions

A few things confuse people when they first meet CDNs. Let’s clear them up:

• “A CDN replaces my server.” No. The CDN sits in front of your origin and caches copies. Your origin is still the source of truth, and the CDN goes back to it on a miss.
• “A CDN can cache everything.” Not really. It shines for static content. Highly dynamic or personalized data should still come from the origin.
• “Once it’s on the CDN, my update shows up instantly.” Nope. The edge keeps the old copy until the TTL expires or you purge it. That’s why you might still see the old version for a while.
• “A CDN is only for huge companies.” Not true. Even a small blog benefits, because faster images and quicker loads help everyone, regardless of size.

🛠️ Design Challenge

Try this one yourself to lock the ideas in.

Imagine Alex is building a photo-sharing site that’s blowing up worldwide. Decide what should go through the CDN and what shouldn’t:

• The uploaded photos and the site’s logo? Static, so cache them at the edge.
• The CSS and JavaScript files? Static, edge them too.
• A user’s private “My Photos” feed and their notifications? Personalized, so keep those going to the origin.

Then ask yourself: when Alex replaces the logo, how do you make sure users stop seeing the old one? Walk through TTL, purging, and the file-renaming trick. This is exactly the kind of reasoning a system design interview is looking for.

🧩 What You’ve Learned

You can now explain how a CDN keeps the web fast at huge scale. Here’s what you’ve picked up:

• ✅ A CDN is a network of edge servers worldwide that cache copies of content near users.
• ✅ The origin server is the source of truth, and edge servers serve cached copies close to users.
• ✅ A cache hit serves from the edge fast, while a miss fetches from the origin and then caches it.
• ✅ CDNs are great for static content like images, CSS, JS, and video, but not for personalized data.
• ✅ They lower latency, ease load on the origin, handle traffic spikes, and add some DDoS protection.
• ✅ You manage freshness with TTL, purging, and renaming files, and you never cache private per-user data.

🧠

Check Your Knowledge

Test what you learned. Pick an answer for each question, then click Check.

1. 1

   What is a CDN?

   A single powerful server that replaces your origin server A network of servers spread across the world that cache copies of content close to users A type of database used to store user accounts A tool that makes your code run faster on one machine
   Check

   Why: A CDN is a worldwide network of edge servers that cache content near users, cutting latency by serving from a nearby server instead of the faraway origin.

2. 2

   What is the difference between a cache hit and a cache miss?

   A hit means the file was deleted, a miss means it was saved A hit serves from the edge right away, a miss fetches from the origin and then saves a copy A hit fetches from the origin, a miss serves from the edge They both mean the origin server is down
   Check

   Why: On a hit the edge already has the file and serves it fast; on a miss it fetches from the origin once, serves it, and caches a copy for next time.

3. 3

   Which kind of content is a CDN best suited for?

   A user's private bank balance A shopping cart that is unique to each user Static content like images, CSS, JavaScript, and video Live data that changes every second per user
   Check

   Why: CDNs shine with static content that is the same for everyone, and are a poor fit for highly dynamic or personalized data.

4. 4

   After you update a file on the origin, why might users still see the old version?

   The CDN permanently deletes all updated files The edge keeps its cached copy until the TTL expires or you purge it Updates can never reach a CDN at all The origin server stops working after any update
   Check

   Why: The edge serves its saved copy until the TTL expires or you purge it, which is why you may still see the old version for a while.

Score: 0 / 4 Reset

🚀 What’s Next?

You’ve got the CDN picture now, so let’s connect it to the bigger story of how the web stays fast.

• Introduction to Caching goes deeper into the caching idea that powers CDNs and a whole lot more.
• What Happens When You Type a URL shows where a CDN fits into the full journey of loading a page.

Get those two down and you’ll have a solid grip on the network and storage fundamentals every system design interview leans on.