A CDN, or Content Delivery Network, has servers around the world that deliver our content to users. Because their servers are closer to our users than our servers are, they can respond faster than we can, improving latency. Here is the way they typically deliver our content:
Then, when a user makes a request, this is what happens the first time:
After the CDN caches it, during the next minute, here is what happens when other users make the same request:
These subsequent requests take much less time to complete than requests to our server because: A) the CDN servers are closer, so it takes less time to reach them over the internet, and B) the CDN servers have the whole response ready to quickly return, whereas our server would spend time constructing the response.