CloudFront CDN - What is Amazon CloudFront?

Objects : Objects are files (for example web pages, images, and digital media files) same as s3, that can be anything that can be served over HTTP. The files must be publicly readable.

Origin Server : An origin server is the location where you store the original, definitive version of your objects. With Amazon CloudFront, your origin server is an Amazon S3 bucket that you own. Using an existing Amazon S3 bucket as your Amazon CloudFront origin server doesn't fundamentally change that bucket in any way; you can still use it as you normally would to store and access Amazon S3 objects (at the normal Amazon S3 prices ).

Distributions : How do you get Amazon CloudFront to recognize them?
For this, you have to create a distribution , which is a link between an Amazon S3 bucket (the origin server) and a domain name (assign by Amazon CloudFront). You use this new domain name in place of standard Amazon S3 references. For example, http://mybucketexplorer.s3.amazonaws.com/image.jpg would instead be http://somedomainname.cloudfront.amazonaws.com/image.jpg.

Note : You can have up to 100 distributions in the Amazon CloudFront system.

Edge Locations : One of the sites Amazon CloudFront uses to cache copies of your content for faster delivery to end users. Amazon CloudFront uses edge locations (United States, Europe, and Asia). When a user requests one of your objects, Amazon CloudFront decides which edge location is best able to serve the request. If the edge location doesn't have a copy, Amazon CloudFront goes to the origin server and puts a copy of the object in the edge location.

Time to Live (TTL) : An object's time to live is the length of time the object can stay in an edge location before it expires. Once the object expires, Amazon CloudFront no longer serves that particular object and need to get a new copy of that object from the origin server to serve to end users.

**Eventual Consistency : When you successfully create, modify, or delete a distribution using the Amazon CloudFront control API, it takes time for your changes to propagate throughout the Amazon CloudFront system. The information about the distribution will eventually be consistent, but an immediate request to the control API to get that distribution's information might not show the change. Consistency is usually reached within minutes, but a high system load or network partition might increase this time. The control API lets you determine when your changes have been fully deployed.


 

Fast : With a network of edge locations around the world, Amazon CloudFront caches copies of your content close to end users, lowering latency for delivery to them. The service also effectively gives you the high, sustained data transfer rates needed to deliver large popular objects to end users at high scale.

Simple : A single API call specifies the Amazon S3 bucket that will be your origin server and lets you get started distributing content through the Amazon CloudFront network.

Designed for Use with Other AWS Services : Amazon CloudFront is tightly integrated with Amazon S3 and works well with other AWS services, like the Amazon Elastic Compute Cloud (Amazon EC2).

Inexpensive : Amazon CloudFront passes on the benefits of Amazon’s scale to you. You pay only for the content that you deliver through the network, without minimum commitments or up-front fees. Because there’s no need to negotiate a contract with a sales person, you can get started in just minutes.

Elastic : With Amazon CloudFront, you don’t need to worry about maintaining expensive web server capacity to meet the demand from potential traffic spikes for popular content. The service automatically responds as demand increases or decreases, without any actions needed from you.

Reliable : Amazon CloudFront is built on Amazon’s highly reliable infrastructure. The distributed nature of the edge locations enables Amazon CloudFront to automatically re-route end users to the next closest location as required by network conditions.

Global : Amazon CloudFront uses a global network of edge locations, located near your end users in the United States, Europe and Asia.


 

How does Amazon CloudFront  delivers you content. See following steps:

1. Place the object in the bucket (make it publicly readable.)
2. Create a distribution and get the domain name (auto generated by Amazon CloudFront).
3. Create object links using the domain name.
4. An end user requests the object and Amazon CloudFront chooses the appropriate edge location.
5. Amazon CloudFront goes to the origin server for the object if necessary.
6. The edge location serves the object.
7. Later, the object expires (TTL).

Origin Server Bucket : When you create the distribution, you specify the origin server bucket's name using this format: <bucket name>.s3.amazonaws.com. Do not use the Amazon S3 path style for specifying the bucket, which is s3.amazonaws.com/<bucket name>.

Caller Reference : The caller reference is a unique value that you provide and Amazon CloudFront uses to prevent accidental replays of your request. Once you create a distribution, you can't change its caller reference.

Enabled or Disabled : When you create a distribution, you need to tell Amazon CloudFront whether you want the distribution to be enabled or disabled once it's created. Enabled means that as soon as the distribution is fully deployed

CNAME Aliases : You can optionally associate one or more CNAME aliases with a distribution so that you can use a domain name of your choice in links to your objects instead of the domain name Amazon CloudFront assigns. You need to be the owner of any domain name that you specify as an alias of your distribution's domain name. You can have up to 10 CNAME aliases per distribution. You can associate a particular CNAME alias with only one distribution.


 

When you create a distribution, you can provide optional comments about the distribution. You can update those comments at any time. The maximum allowed length is 128 characters.