How Does A Content Delivery Network Make Your Site Faster?

Content Delivery NetworkYour site’s files are stored on a server in a data center. When a browser requests a page, it connects to the web server — usually via lots of other machines like routers and switches — which will send the browser the necessary files. If the browser is close to the server, it won’t take long for the files to traverse the internet. If the browser is on a computer on the other side of the world — or even just the other side of the country — it might take a relatively long time. The browser has to send requests to the server, and the server has to send data to the browser, a process repeated many times for every page request.

As you might imagine, the further apart the browser and the server are, the longer it takes to get the files. For long distances, the latency can be considerable, creating a less than ideal experience for a visitor or shopper.

One of the jobs of a content delivery network is to reduce that latency to an acceptable level. A CDN is a network of servers in lots of different data centers all over the world. Your site’s files are replicated on each of these servers, and when visitors request a page from your site, they are redirected to the nearest server in the CDN. The CDN’s servers are called edge nodes.

There are many different ways to get content to the edge nodes, but I’d like to discuss the two techniques that the Future Hosting content delivery network uses.

Origin Pull

The origin pull system works much like a traditional cache, except the data is cached in the edge nodes rather than near the origin server.

With origin pull, the files are uploaded to, and remain on your hosting account’s server. When a distant visitor requests a page, they’re sent to their nearest edge node. If the assets of the page aren’t already on the edge node, it will make a request to your server and download the files, before sending them on to the visitor’s browser.

The files will then be cached on the edge node for a predetermined amount of time, usually 24 hours. Next time someone tries to visit the same page, the data can be sent to them from the edge node without your server being involved.

PoP Pull

A PoP Pull CDN work much as we’ve already discussed, except that the files are uploaded directly to the CDN by the user with FTP or a web-based control panel. PoP Pull CDNs can be more efficient, but they’re not automatic and transparent to the client in the same way a Origin Pull CDN is.

Synchronization

Depending on the specific implementation, synchronization of the edge nodes is handled differently for Origin Push and PoP Push CDNs. In the former, content is not automatically synchronized between edge nodes — it’s only sent to an edge node when it’s requested by a browser. PoP Push systems automatically synchronize edge-nodes after the content is manually uploaded. In practice, for most sites the difference in performance is minimal — the first user to visit a specific edge-node will experience slower performance with Orign Push as the data is fetched from the origin server, but after that’s done, performance at each node is equivalent.

Dynamic Or Static

I’ve only explained the simplest case here. Static content can be sent to edge nodes without complication, but if a site uses a dynamic content management system like WordPress or an eCommerce store, often there isn’t a “page” to be cached in an edge node because pages are generated individually for visitors. In that case, the origin server will be doing most of the work. However, even dynamic sites have lots of static assets like CSS and JavaScript files, and images. Static assets like these are perfectly compatible with a CDN.

Matthew Davis is a technical writer and Linux geek for Future Hosting.

Dedicated Server Special

Take advantage of our Double RAM offer on the E3-1230v2 4 x 3.30GHz+HT server! Only $134.95 per month. Managed and Unmanaged options available at checkout.

GET STARTED