AWS storage services and replication strategies (for example Amazon S3, Amazon RDS, Amazon ElastiCache)

What is Amazon S3 and why would you use it over EBS for storing files?

Amazon Simple Storage Service (S3) is an object storage service that offers scalability, availability, security, and performance. You would use it over Amazon Elastic Block Store (EBS) for storing files that need to be accessed over the Internet because S3 is designed for high durability and accessibility across the Internet. S3 also offers features like versioning, lifecycle policies, and event notifications that are not available with EBS.

Can you explain the difference between S3 Standard, S3 Standard-Infrequent Access (S3 Standard-IA), and S3 One Zone-Infrequent Access (S3 One Zone-IA)?

S3 Standard is designed for frequently accessed data, providing high availability and durability. S3 Standard-IA is for less frequently accessed data, offering a lower storage price but with a retrieval fee. S3 One Zone-IA is also for infrequently accessed data but stores data in a single Availability Zone, which reduces the cost further but comes with a higher risk of data loss if that zone is compromised.

How does Amazon RDS facilitate database replication and what replication options does it provide?

Amazon Relational Database Service (RDS) provides replication features to enhance database availability and reliability. RDS supports two types of replication: Multi-AZ deployments for high availability, which automatically provisions and maintains a synchronous standby replica in a different Availability Zone; and read replicas for scalability, providing asynchronous replication to create read-only copies of the database in the same or different regions.

What would be a reason to use RDS Multi-AZ deployments over RDS read replicas?

RDS Multi-AZ deployments are used for high availability purposes; they automatically fail over to the standby in case of an outage, minimizing downtime. Read replicas, on the other hand, are used primarily for scaling read operations and do not provide automatic failover functionality.

Describe the use case for Amazon ElastiCache and when it would be appropriate to implement it in an AWS architecture.

Amazon ElastiCache is an in-memory caching service designed to improve the performance of web applications by allowing you to retrieve information from fast, managed, in-memory caches, rather than relying solely on slower disk-based databases. It’s appropriate to use ElastiCache when you need high throughput and low latency for read-heavy application workloads or compute-intensive workloads.

What are the main differences between Amazon S3 and Amazon EFS?

Amazon S3 is an object storage service suitable for a wide range of storage needs but is not natively accessible as a file system by EC2 instances. Amazon Elastic File System (EFS) is a fully-managed file storage service that provides simple, scalable file storage for use with Amazon Cloud services and on-premises resources. EFS is accessible to multiple EC2 instances simultaneously using a file system interface and file system semantics.

How do you secure data at rest in Amazon S3?

To secure data at rest in Amazon S3, use S3 server-side encryption (SSE) which offers three key management options: SSE-S3 (using AWS managed keys), SSE-KMS (using AWS KMS customer master keys), and SSE-C (using customer-provided keys). Access controls like IAM policies, bucket policies, and access control lists (ACLs) further enhance data security by defining who can access S3 resources.

What strategies can be used to reduce data transfer costs with Amazon S3?

To reduce data transfer costs in Amazon S3, one could use S3 Intelligent-Tiering to optimize storage costs automatically, cache content using Amazon CloudFront, compress data to reduce size, use S3 Transfer Acceleration for faster uploads across distances instead of over the public internet, and select the appropriate S3 storage class based on access patterns.

How would you perform a disaster recovery drill on an Amazon RDS database?

To perform a disaster recovery drill on an RDS database, you can manually create a snapshot of your database, then restore it to a new instance in a different region. It’s important to test failover to a standby instance in Multi-AZ deployments, and also to promote a read replica to a standalone instance if you’re relying on read replicas for disaster recovery.

Can you set up cross-region replication for Amazon S3? If yes, how?

Yes, cross-region replication (CRR) can be setup for Amazon S3 by enabling it in the bucket settings. You need two buckets in different regions, where one serves as the source and the other as the destination. You also need to specify an IAM role that Amazon S3 can assume to replicate objects on your behalf.

In what scenario would you recommend using AWS Snowball, and what is it?

AWS Snowball is a data transport solution that utilizes secure, shippable storage devices to transfer large amounts of data into and out of AWS. It is recommended in scenarios where transferring data over the internet is too slow, costly, or where network connectivity is insufficient. This solution is ideal for data migration or data transport needs in scenarios like large-scale data centers migrations, disaster recovery, or collecting and processing data in edge locations.