RDS Flashcards

This deck aims to help retain concepts related to the AWS RDS service.

1
Q

What AWS service simplifies the setup, operation, and scaling of a relational database in the cloud?

A

Amazon Relational Database Service (RDS)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Can you access the operating system of RDS instances?

A

No, RDS is a managed service that does not provide access to the operating system or allow SSH access

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Which database engines are supported by RDS?

A

Amazon Aurora, MySQL, MariaDB, PostgreSQL, Oracle, and Microsoft SQL Server

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What type of replication is used when data is replicated to a standby instance?

A

Synchronous replication, ensuring that both the primary and standby instances have identical data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Where are RDS instances hosted?

A

RDS instances are hosted within a VPC in a specific AWS region, each instance has dedicated storage provided by EBS, and deployments can include multiple databases

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Where are RDS backups stored?

A

Both automated backups and manual snapshots are stored in Amazon S3 (AWS-managed bucket)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Why is running and managing a database directly on an EC2 instance generally considered a bad practice?

A

It involves significant administrative overhead and risk
Managing backups and disaster recovery is complex
EC2 and EBS are limited to a single availability zone
Lacks advanced features and performance enhancements offered by AWS database products
Does not support serverless options, easy scaling, or replication

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What type of replication is used when data is replicated to a read replica, typically located in another region?

A

Asynchronous replication

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What are the billing components of the RDS service?

A

Instance size and type
Multi-AZ deployment (additional instances)
Monthly storage fee per GB
Data transfer cost per GB (inbound and outbound from the internet or other AWS regions)
Backup and snapshot storage per GB (snapshot storage is free up to the amount of provisioned storage, e.g., 2TB of storage equals 2TB of free snapshot storage)
Licensing fees for commercial database engines (if applicable)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

When might you consider running a database directly on an EC2 instance instead of using RDS?

A

When access to the database instance’s operating system is required
When advanced database tuning options are needed
When the database engine or version is not supported by RDS
When a specific OS or database combination that AWS does not provide is required
When a particular architecture not supported by AWS is necessary (e.g., specific replication or resilience requirements)
When there are vendor-specific requirements
When decision-makers have a strong preference for direct control

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What feature of RDS automatically creates a standby database instance, synchronously replicates data to it in a different availability zone, and provides automatic failover without manual intervention?

A

RDS Multi-AZ

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What are the most popular RDS Multi-AZ architectures?

A

Instance Multi-AZ architecture (primary-secondary)
Cluster Multi-AZ architecture (writer-reader-reader)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Based on the following characteristics, identify the Multi-AZ architecture:

  • Synchronous data replication
  • Only one standby replica
  • Standby replica cannot be used for reads or writes until failover
  • Failover takes 60-120 seconds (DNS update)
  • Located in the same region (different AZ)
  • Backups can be taken from the standby replica to improve performance
  • Provides one endpoint that points to either the primary instance or the standby instance during failover
A

Instance Multi-AZ architecture (primary-secondary)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Based on the following characteristics, identify the Multi-AZ architecture:

  • One writer synchronously replicates to readers in different AZs
  • Two readers only
  • Primary instance is used for write operations only
  • Reader instances are used for read operations only
  • Utilizes high-performance hardware (e.g., Graviton + local NVMe SSD storage)
  • More efficient replication via transaction logs
  • Writes are committed when at least one reader confirms replication
  • Two endpoints provided:
    • Cluster endpoint for primary instance (writer), used for reads, writes, and administration
    • Reader endpoint for available readers, used for reads (may point to the primary instance for reads)
    • Instance endpoints for specific instance access, generally for testing and troubleshooting
A

Cluster Multi-AZ architecture (writer-reader-reader)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What might trigger an RDS failover event?

A

Availability Zone outage
Primary instance failure
Manual failover
Instance type change
Software patching

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What backup and restore services are available within AWS RDS?

A

Automated Backups and Manual Snapshots

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Based on the following characteristics, identify the RDS backup type:

  • Automated, snapshots taken daily
  • Retention period from 0 to 35 days
  • Stores transaction logs every 5 minutes
  • Can be deleted with RDS or retained until the expiration of the retention period
  • Cross-region replication can be configured (not default)
  • Allows recovery to any 5-minute point in time using the latest snapshots and transaction logs
A

Automated Backup

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Based on the following characteristics, identify the RDS backup type:

  • Not automatic
  • Persist after RDS instance termination (must be manually deleted)
  • Represents a single point in time
A

Manual Snapshot

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Are automated backups or manual snapshots restored directly to the current RDS instance?

A

No, restoring from backups or snapshots creates a new RDS instance, resulting in a new database endpoint, and the restoration process can be slow

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Which RDS feature allows the creation of one or more read-only copies of database instances within the same AWS region or across different regions to enhance scalability?

A

RDS Read Replica (improves global availability and resiliency)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Should applications using RDS be aware of read replicas?

A

Yes, applications should be aware of read replicas as they have their own endpoint addresses

21
Q

How many direct read replicas can an RDS instance have?

A

Up to 5 read replicas per database instance, read replicas can also have their own read replicas, though replication lag may occur

22
Q

Can RDS read replicas be used for write operations?

A

Only after a read replica has been promoted to a primary instance

23
Q

How do RDS snapshots and read replicas impact RPO (Recovery Point Objective) and RTO (Recovery Time Objective)?

A

Snapshots and backups improve RPO, as they allow recovery to a specific point in time, however, they do not directly impact RTO, as restoring from backups can be time-consuming
Read replicas offer near-zero RPO and can be quickly promoted to a primary instance, resulting in a low RTO if recovering from failure, but not from data corruption

24
Q

What data security features are available in RDS?

A

Authentication
Authorization: Controlled by the internal database engine (IAM is used to authenticate, not authorize).
Encryption in Transit: Data between the client and RDS instance is encrypted using SSL/TLS (can be mandatory).
Encryption at Rest: EBS + KMS encryption, handled by the host; the database engine is unaware of encryption (default)

25
Q

Which components of Amazon RDS are encrypted?

A

Storage, logs, snapshots, and replicas are encrypted using the same customer master key, once encryption is enabled, it cannot be removed

26
Q

Which RDS-supported database engines offer support for Transparent Data Encryption (TDE)?

A

MSSQL and Oracle, encryption is handled by the database engine (not the host), additionally, Oracle supports TDE integration with CloudHSM, ensuring no key exposure to AWS

27
Q

Which managed database service is designed for applications that require customization of the underlying operating system and database environment?

A

RDS Custom

28
Q

Which database engines are supported by RDS Custom?

A

RDS Custom

29
Q

What is the maximum storage capacity that RDS (Aurora provisioned) can have?

A

Aurora provisioned uses shared cluster volume (instead of local storage) available for all compute instances within a cluster, with a maximum of 128 TiB. This includes:
- SSD high IOPS low latency storage nodes
- Storage nodes accessible by all compute instances.
- The primary instance can write to the storage node, and replica and primary instances can read
- Automatic repair of storage nodes
- Adding or removing replicas without storage provisioning or removal
- Storage is billed based on consumption

29
Q

What are the primary benefits of using RDS Custom?

A

Provides the benefits of RDS automation with the access needed for legacy, packaged, and custom applications
Enables connection through SSH, RDP, or Session Manager
Provides limited access to the Operating System and Database Engine

30
Q

What is the maximum number of read replicas that can be created with RDS (Aurora provisioned)?

A

Aurora provisioned supports a cluster consisting of a single primary DB and 0-15 replicas, which can be used for both availability and read operations

31
Q

What endpoints are exposed when using RDS (Aurora provisioned)?

A

Cluster endpoint (pointing to the primary instance)
Reader endpoint (balancing load between replicas and the primary instance)

32
Q

What primary benefits does Amazon Aurora offer when used with RDS (Aurora provisioned)?

A

Faster provisioning, improved availability, and better performance

33
Q

What are the primary considerations when using Amazon RDS (Aurora provisioned)?

A

No free tier option
Does not support micro instances
Compute hourly charge per second, with a 10-minute minimum
Charged for storage GB/month consumption and IO cost per request
100% DB size in the backup is included for free

34
Q

When restoring an RDS (Aurora provisioned) instance from a snapshot, is the restoration performed in the current database?

A

No, restoring from snapshot a new cluster is created

35
Q

What feature of RDS (Aurora provisioned) enables in-place rewind?

A

The Backtrack feature enables rollback in place to a point in time (before data corruption occurred)

36
Q

What feature of RDS (Aurora provisioned) enables much faster database creation?

A

The Fast Clone feature allows for much faster database creation than copying all the data

37
Q

What are the primary features of Amazon Aurora Serverless?

A

Aurora Serverless cluster has MIN and MAX Aurora Capacity Units (ACUs)
ACUs are allocated from a shared pool managed by AWS
The cluster adjusts based on load, can scale to 0, and be paused
Consumption is billed on a per-second basis
Connections are managed through a shared proxy fleet managed by AWS

38
Q

What is the ideal use case for Amazon Aurora Serverless?

A

Infrequently used applications, new applications, test and development workloads, variable/unpredictable workloads, and multi-tenant applications

39
Q

Which Aurora deployment consists of a primary Aurora DB Cluster in one AWS Region, where data is written, and read-only secondary DB clusters in other AWS Regions?

A

Aurora Global Database enables database reads scaling across the world and places applications close to users

40
Q

What are the primary considerations when using the Aurora Global Database?

A

Consists of a primary instance and up to 15 read replicas in the primary region
Up to 16 read replicas in the secondary region, with up to 5 secondary regions
Read replicas can be promoted to primary
The cluster volume has approximately 1-second replication time between regions without impacting DB performance (as it occurs at the storage layer)
Enables cross-region Disaster Recovery (DR) and Business Continuity (BC)

41
Q

Which feature enables the creation of several read-write instances of the Aurora database across multiple AZs?

A

Multi-master writes

42
Q

What are the primary considerations when using the Aurora Multi-master writes feature?

A

Created instances are capable of both read and write operations
There is no concept of a load-balanced endpoint; the application manages which instance to connect to
Data is committed to cluster-shared storage (all nodes) and replicated to the other instances to update their in-memory cache
During a failover event, switching to another writer is instantaneous

43
Q

Which RDS feature represents a fully managed, highly available database proxy that makes applications more scalable, resilient to database failures, and more secure?

A

RDS Proxy

44
Q

What are the primary features of RDS Proxy?

A

Fully managed DB proxy for RDS and Aurora
Built-in auto-scaling and high-availability features
Maintains a long-term connections pool
Applications connect to the proxy rather than directly to the database
Accessible only from VPC through a proxy endpoint
Can enforce SSL/TLS connections
Reduces failover time by over 60%
Abstracts database failure from the application

45
Q

What service allows for zero data loss and low or zero downtime migrations between two database endpoints, moving databases into or out of AWS?

A

Database Migration Service (DMS)

46
Q

Can AWS DMS migrate data between two on-premises databases?

A

No, at least one endpoint must be on AWS

47
Q

How does the AWS DMS service operate?

A

It runs a replication instance that executes one or more replication tasks and requires source and destination endpoints, which point to the source and target databases

48
Q

What types of migration jobs are available in AWS DMS?

A

Full load: One-off data migration (if an outage is affordable)
Full load + Change Data Capture (CDC): Migrates data and replicates ongoing changes
Change Data Capture (CDC) only: Alternative method to transfer bulk DB data using native tooling and then use CDC to replicate changes

49
Q

Which tool assists with schema conversion between different database versions or engines when using AWS DMS?

A

Schema Conversion Tool (SCT), can be used to migrate database data to S3 and is NOT used when migrating between compatible database engines

50
Q

Can AWS DMS be used with a range of Snow products?

A

Yes, a common use case for multi-TB database migrations:
1. Use SCT to extract data from the database and store it in the Snowball
2. Ship the Snowball to AWS, where the data is loaded onto an S3 bucket
3. Use DMS to migrate data from S3 to the target store
4. Use CDC to update S3 via the internet with ongoing changes