Highly Available Datastores – High Availability and Resilience – SOA-C02 Study Guide

Highly Available Datastores

This chapter covers the following official AWS Certified SysOps Administrator – Associate (SOA-C02) exam domain:

Domain 2: Reliability and Business Continuity

CramSaver

If you can correctly answer these questions before going through this section, save time by skimming the Exam Alerts in this section and then completing the Cram Quiz at the end of the section.

1. True or False: To make an S3 bucket on a standard tier highly available, you must ensure it is replicated across at least one more availability zone.

2. True or False: Moving an object from S3 Standard to S3 Infrequent Access requires you to download and delete the object from S344 Standard and upload it to S3 Infrequent Access.

Answers

1. Answer: False. The S3 Standard tier is automatically replicated across at least three availability zones.

2. Answer: False. The object can be life-cycled based on time, or intelligent tiering can be used to move the object automatically.

Amazon S3

Amazon S3 is a fully managed, serverless object storage service accessible via HTTP and HTTPS. The service has a 99.99 percent regional high-availability SLA and guarantees 99.999999999 percent of data durability. In more human terms, the durability SLA means there is a probability of losing 1 file out of 10 million every 10,000 years.

Any content stored on S3 must be stored within a bucket. Each bucket serves as a unique endpoint for objects, and each object has a unique key within the bucket. Each key is composed of the filename and one or more prefixes. Prefixes can be used to structure the files even further and to represent a directory-like view of the files because S3 has no concept of directories. Two types of URLs can be used to access an S3 bucket:

Virtual-hosted-style URLs (current)

Path-style URLs (currently expecting deprecation)

Virtual-hosted-style URLs are now the default way of accessing S3 buckets and have the following format:

Click here to view code image

{bucket-name}.s3.{region-id}.amazonaws.com}/{optional key prefix}/

{key-name}

In this case the bucket name is a subdomain of s3.{region-id}.amazonaws.com, which means you can create a CNAME record in your DNS service that can point directly to the S3 bucket.

ExamAlert

The S3 service previously supported a global S3 region virtual-hosted-style URL where you were able to omit the region parameter completely, and the format allowed requests to bucket URLs with the following format:

Click here to view code image

{bucket-name}.s3.amazonaws.com

Any buckets created after March 20, 2019, do not support the legacy global endpoint anymore.

The path-style URL is currently being deprecated. Deprecation was scheduled for September 30, 2020, but due to client requests, AWS has delayed the deprecation. Any buckets created after September 30, 2020, cannot be accessed via path-style URLs. The structure of a path-style URL is as follows:

Click here to view code image

http{s}://s3.{region-id}.amazonaws.com/{bucket-name}/

{optional key prefix}/{key-name}