Cloud Storage (AWS S3 Educate Note)

date

Apr 29, 2023

slug

AWS_C1

author

status

Public

Intro

Nowadays, cloud services have become increasingly common. Hosting servers locally is no longer the only option. Companies can rent a cloud server flexibly, depending on floating usage, and reduce maintenance fees. Additionally, programmers can focus on features without worrying about deployment. Since it is a common and useful approach, I would like to record my understanding while going through the AWS Educate course to prevent the knowledge from slipping out of my mind.

Structure of S3

AWS S3 is an object storage, if you are not familiar with the deference between three types of storage: block storage, file storage, and object storage. Check the RedHat for details.

Objects

In S3, objects are saved into buckets and placed into regions. Objects consist of three main parts: data, metadata, and key. The data is the content, while the metadata contains information about the data, such as the content type and last modified date. The key serves as an identifier.

Buckets

A bucket is a container of objects that has six different types. Each type has its own properties and different prices. When learning about S3, the most important consideration is choosing the correct bucket type that is both cost-effective and meets the requirements.

The price of data storage is typically proportional to its latency. Therefore, if the data is not frequently used, it can be saved in Glacier Flexible Retrieval or Deep Archive. The most impressive option is Intelligent-Tiering, which can handle cases where latency is hard to predict.

For example:

If a company wants to host a static website, lower latency is necessary, and the standard option would be the best choice.

In a data center that needs to backup data for the long term and does not need frequent access, Glacier Flexible Retrieval or Deep Archive might be the better options.

Regions

AWS has data centers located all around the world, allowing you to choose where to put your data. Generally, the closer the region is to your customer, the better the performance will be.

Conclusion

AWS S3 is an object storage service that saves objects into buckets placed into regions. Objects consist of data, metadata, and a key. Buckets have six types, each with different properties and prices. Choosing the correct bucket type is important for cost-effectiveness and meeting requirements. AWS has data centers located worldwide, allowing for data placement optimization.