S3

Question 1

S3

Answer 1

• Global public service
• Data hosted in a specific region, and auto replicated across multiple AZ’s
• Object store. Must replace object to modify file
• By default, all S3 buckets are private and only accessible to that account

Question 2

S3 Limitations

Answer 2

• CANNOT mount S3 to instances
• Max object size is 5TB
• Cannot upload more than 5Gb at a time (need to use multi part upload)
• Bucket name must be globally unique
• 100 buckets per account (soft limit), 1000 buckets hard limit
• ## Cant create buckets for users (per above limitation)

Question 3

Cross Origin Resource Sharing (CORS)

Answer 3

Bucket configuration to allow client web applications on one domain to request for objects in another domain.

Question 4

S3 MFA-Delete

Answer 4

Requires a root user to configure MFA-Delete setting. This means users must perform MFA validation to permanently delete an object version, and/or suspend versioning on the bucket.

Question 5

S3 Cross Region Replication

Answer 5

Asynchornous replication to another bucket. MUST have versioning enabled.
Use cases: compliance, lower latency access, replication across accounts

Question 6

S3 Lifecycle Configuration

Answer 6

Transition Actions: change storage class of data
Expiration Actions: expire objects or versions of objects.
Minimum of 30 days before transition

Question 7

S3 Select and Glacier Select

Answer 7

Getting specific set of data by using SQL and server side filtering to reduce data download and time taken

Question 8

Origin Access Identity

Answer 8

You can configure an S3 bucket as the origin of a CloudFront distribution. OAI prevents users from viewing your S3 files by simply using the direct URL for the file. Instead, they would need to access it through a CloudFront URL.