Hello!

I’ve been preparing for the Amazon Web Services (AWS) Solutions Architect certification and recently learned about the S3 storage service, which I will be exploring in today’s blog in detail.

What is Amazon S3?

Amazon S3, short for Amazon Simple Storage Service is an object-based cloud storage service provided by Amazon under the Amazon Web Services (AWS) family of cloud products.

The S3 service aims to provide the following gurantees:

  1. Secure
  2. Highly available
  3. Scalable
  4. High performance
  5. Durability (99.999999999% - Amazon’s 11 9’s promise!)

Some specifics

  • Files can range from 0 bytes to 5 terabytes in size.
  • Virtually unlimited storage
  • Collections of files are stored in objects called Buckets
  • All buckets on the S3 platform exist in a single global namespace and hence each bucket should have a unique name, globally.

  • A unique URL is generated against each bucket.

    [Image source: Amazon.com]

Data consistency model

The Amazon S3 platform provides the following two data consistency gurantees:

  1. Read after Write consistency for newly-uploaded files.

    For example, if I upload a new file titled test.txt to an empty bucket, and immediately try to access test.txt, I will be able to access the file with no issues.

  2. Eventual consistency for updates and deletes.

    For example, I now make an edit to the previously-uploaded test.txt and re-upload it to my bucket. Now, when I try to immediately access test.txt, it is not guranteed what version I will be served, it may either be the old version or the new. However, eventually, I will always be served the latest version of test.txt.

    Similarly, when I delete test.txt from the bucket, it may still be stored on the S3 servers for some time, however, eventually, the file will be deleted.

Features

  1. Versioning - allows multiple versions of the same object to be stored in the S3 bucket.

  2. Static website hosting - the S3 service allows hosting static HTML web pages with no support for server-side scripting for dynamic content such as PHP.

  3. Access Control Lists (ACLs) - provide control of access to specific files within a bucket.

  4. Bucket Policy - similar to ACLs, however these implement bucket-wide policies as compared to object-specific policies in case of ACLs.

  5. Server-side encryption - controls whether data within S3 servers is encrypted before storage.

Demo and Tutorial

Creating an Amazon S3 bucket

  1. From the AWS Services page navigate to S3 under the Storage category.

  2. Click Create bucket.

  3. In the Create bucket page, do the following:

    i. Choose a unique Bucket name.

    ii. Select the Region where you would like the bucket to be stored.

    iii. Configure access to bucket via ACLs or Bucket Policy.

  4. Finally click Create bucket.

  5. The bucket is created:

Enabling bucket access logging

  1. Repeat the bucket creation procedure described in the previous section to create a bucket for log storage.

  2. From the bucket homepage, navigate to Properties

  3. Click on the Server access logging tile.

  4. Click Enable logging.

  5. In the Target bucket section choose the bucket you created in Step 1.

  6. In the Target prefix section choose a prefix that you want to prepend to all log files for easy filtering.

  7. Finally press save!

Enabling object versioning

  1. From the bucket homepage, navigate to Properties.

  2. Click on the Versioning tile.

  3. Click Enable versioning.

  4. Click Save.

  5. To see versioning in effect lets upload an empty text file titled test.txt.

  6. Now let’s write Hello World in test.txt and re-upload it our bucket.

  7. In order to browse the two versions of test.txt, do one of the following:

    i. Single click the file, click Latest version, and select the desired version from the drop-down menu and press Download.

    ii. Switch the Versions slider to Show, you will now see two entries under test.txt, download whichever version is required.

Static Website Hosting

  1. Let’s first create an HTML file titled index.html with the following contents:

     <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
 <html xmlns="http://www.w3.org/1999/xhtml">
     <h1>
         Hello world!
     </h1>
 </html>

  2. Upload index.html to the bucket.

  3. From the bucket homepage, navigate to Properties.

  4. Check the Use this bucket to host a website combo box.

  5. In the Index document field, select index.html.

  6. In the Error document field, select an HTML file that you want to serve in case of an error, if you did not create an error file, select index.html.

  7. Note down the Endpoint URL.

  8. Click Save.

  9. Visit the endpoint URL to view the hosted website!