WTF is Amazon Simple Storage Service (Amazon S3)?

Tomasz Łakomy: [0:00] According to the Amazon S3 landing page, Amazon S3, also known as Simple Storage Service, it is an object storage service that offers industry-leading scalability, data availability, security, and performance.

[0:13] We can also see that Amazon S3 provides easy-to-use management features, and it is also designed for 99.999, 11 nines of durability, and it stores data for millions of applications for companies around the world.

[0:29] There are a couple of things to notice here. First off, Amazon S3 is object-based, which means that it can be used to store files. For instance, it is not suitable to install an operating system on. Do not try that.

[0:42] It also offers, honestly, a ridiculously high durability of your files that is also known as 11 nines because there are 11 nines here. In other words, whatever you will put into S3, you can safely assume that it is going to be safely stored there.

[0:58] Also, this part about storing data for millions of applications, this is honestly 100 percent true. Even if you've never used S3 before directly, it is likely that you have seen an URL like this that is going to end with s3.somethingsomething.amazonaws.com.

[1:15] For instance, this is a photo of me, I'm the one on the right, which is publicly available, and it is hosted on Amazon S3, but where exactly is this photo? How does Amazon S3 store files?

[1:28] Amazon S3 stores files in things that are called buckets. Buckets are containers for data stored in S3. Here we can see the list of buckets I have on my personal account. The S3 itself is a global service, which means that it does not require region selection, but each one of those buckets is actually hosted in a certain AWS region.

[1:48] I am based in Europe. Most of my buckets are in Europe, like Frankfurt or Ireland, but you also have some buckets in North Virginia. There's one very important thing to remember about S3 buckets. Bucket names are globally unique.

[2:01] What that means is if I have a bucket on my account, which is called half-life-3-source-code, it is not possible for any of you to create a bucket with the exact same name. Even if you wanted to create it in a different region, you will not be able to do that because the bucket names have to be globally unique.

[2:20] Because of that, it is a somewhat common practice to add some random characters at the end of a bucket name to ensure that the name is going to be unique. Of course, the content of the bucket doesn't have to be unique, so you can put whatever you want in there.

[2:34] Bear in mind that, by default, S3 buckets are going to be private, and you have to explicitly decide to make a bucket public because for instance, if you want to host a website, you'll probably want to make the index.html file public, but if you're going to store some HR documents, you probably don't want to have it accessible from the Internet.

[2:54] The file size of things that you are going to store in S3 can be from bytes to 5 TB. Honestly, I have to admit that I am yet to see a 5 TB file, but you may have a use case for that.

[3:06] Apart from that, S3 itself offers unlimited storage. In that context, unlimited means literally whatever you can afford.

[3:14] Speaking of money, Amazon S3 is included in AWS Free Tier, so for 12 months, you can get 5 GB of standard storage for free as well as 20,000 GET requests that is getting data from an S3 bucket and also 2,000 PUT requests that is putting data into the bucket.