AWS Certified Solutions Architect Design Resilient Architectures Syllabus

Define RTO (Recovery Time Objective).

RTO is the maximum acceptable length of time an application can be down (unavailable) after a failure or disaster before it must be restored—i.e., the target time to recover service.

Define RPO (Recovery Point Objective).

RPO is the maximum acceptable amount of data loss measured in time—how far back in time recovery occurs. An RPO of 1 hour means up to 1 hour of data may be lost, dictating backup/replication frequency.

List the four AWS disaster recovery strategies in order of increasing cost and decreasing RTO/RPO.

1) Backup and Restore (highest RTO/RPO, lowest cost), 2) Pilot Light, 3) Warm Standby, 4) Multi-Site Active/Active (lowest RTO/RPO, highest cost).

Describe the Pilot Light DR strategy.

Core/critical components (e.g., databases with replication) run continuously in the recovery Region at minimal scale, while application servers are switched off or not provisioned. On disaster, you start/scale the remaining infrastructure. Lower RTO than backup/restore, moderate cost.

Describe the Warm Standby DR strategy.

A scaled-down but fully functional copy of the production environment always runs in the recovery Region. On disaster you scale it up to full capacity. RTO/RPO are lower than pilot light because the system is already running, at higher cost.

Describe the Multi-Site Active/Active DR strategy.

Full production workloads run simultaneously in multiple Regions serving live traffic (e.g., via Route 53 routing). Provides near-zero RTO/RPO but is the most complex and expensive strategy.

Which DR strategy has the lowest cost but the highest RTO/RPO?

Backup and Restore—data and configurations are backed up (e.g., to S3) and restored after a disaster. Cheapest to maintain, but recovery takes the longest because resources must be provisioned from scratch.

What AWS service provides centralized, policy-based backup across multiple services (EBS, RDS, DynamoDB, EFS, etc.)?

AWS Backup. It centralizes and automates backups across AWS services using backup plans, schedules, lifecycle rules, cross-Region/cross-account copy, and Backup Vault Lock for immutability.

What is the difference between an EBS snapshot's first snapshot and subsequent snapshots?

EBS snapshots are incremental: the first snapshot copies all written blocks, and each subsequent snapshot saves only blocks changed since the last snapshot. Snapshots are stored in S3 (managed) and are Region-scoped but can be copied across Regions.

How do you replicate an S3 object to another Region, and what feature enables it?

Use Cross-Region Replication (CRR), part of S3 Replication. It requires versioning enabled on both source and destination buckets and an IAM role. CRR asynchronously copies objects to a bucket in a different Region for DR/latency. Same-Region uses SRR.

What is Amazon Aurora Global Database and what RPO/RTO does it target?

Aurora Global Database replicates a primary Region's database to up to 5 secondary Regions with typical replication lag under 1 second. It targets RPO of ~1 second and RTO under 1 minute via cross-Region failover, used for low-latency global reads and DR.

What is DynamoDB Global Tables?

A multi-Region, multi-active replication feature where DynamoDB tables are automatically replicated across chosen Regions with last-writer-wins conflict resolution, enabling local low-latency reads/writes and Regional failover (near-zero RTO/RPO).

What is Amazon Route 53 health checking, and how does it enable failover?

Route 53 monitors endpoint health (HTTP/HTTPS/TCP) and, with Failover routing policy, returns the primary record while healthy and automatically switches DNS responses to a secondary (standby) endpoint when the primary fails.

Name three Route 53 routing policies useful for resilience and what each does.

Failover (active-passive switch on health check failure), Latency-based (route to lowest-latency Region), and Weighted (split traffic by assigned weights, useful for blue/green and gradual cutover). (Also Geolocation and Multivalue answer.)

What does Route 53 Multivalue Answer routing provide?

It returns up to 8 healthy records selected at random in response to DNS queries, with health checks, giving simple client-side load distribution and improved availability without a load balancer.

What is AWS Elastic Disaster Recovery (DRS) used for?

AWS Elastic Disaster Recovery continuously replicates on-premises or cloud servers (block-level) into a low-cost staging area in AWS, enabling fast, automated failover and failback with minimal RPO (seconds) and RTO (minutes) for full-server recovery.

How does automated failover work in an Auto Scaling group across AZs?

If an instance or an entire AZ fails, the ASG's health checks mark instances unhealthy, terminate them, and launch replacements in healthy AZs to maintain desired capacity, automatically rebalancing across AZs—self-healing without manual intervention.

What CloudWatch construct triggers automated scaling or recovery actions?

A CloudWatch Alarm. It watches a metric against a threshold over defined periods and triggers actions such as Auto Scaling policies, SNS notifications, or EC2 actions (reboot/recover/stop/terminate) for automated failover/remediation.

What does EC2 Auto Recovery (the recover action) do?

For supported instances, a CloudWatch alarm on a system-status-check failure can automatically recover the instance—restarting it on new underlying hardware while preserving instance ID, private/Elastic IP, and EBS volumes.

What is the durability guarantee of Amazon S3 Standard storage?

S3 is designed for 99.999999999% (11 nines) durability of objects over a given year, achieved by automatically storing data redundantly across multiple devices in a minimum of three Availability Zones.

How many AZs does S3 store data across, and which storage class is the exception?

Most S3 storage classes (Standard, Standard-IA, Intelligent-Tiering, Glacier) store data redundantly across at least three AZs. S3 One Zone-IA is the exception—it stores data in a single AZ (lower cost, lower resilience).

What type of storage is Amazon EBS, and what is its key resilience characteristic?

EBS provides persistent block storage volumes attached to a single EC2 instance. Each volume is automatically replicated within its Availability Zone to protect against component failure; it is AZ-scoped (not multi-AZ) and durable independent of instance life.

Compare EBS gp3 vs io2 Block Express volume types.

gp3 is general-purpose SSD with baseline 3,000 IOPS / 125 MB/s (independently scalable up to 16,000 IOPS). io2 Block Express is provisioned-IOPS SSD for the highest performance (up to 256,000 IOPS, sub-millisecond latency, 99.999% durability) for critical databases.

Can an EBS volume be attached to multiple EC2 instances, and how?

By default an EBS volume attaches to one instance in the same AZ. With EBS Multi-Attach (io1/io2), a single volume can be attached to up to 16 Nitro-based instances in the same AZ concurrently (requires a cluster-aware filesystem).

What is the difference between block storage (EBS), object storage (S3), and file storage (EFS)?

Block storage (EBS) presents raw volumes for a single instance—low latency, like a disk. Object storage (S3) stores objects with metadata, accessed via API, infinitely scalable. File storage (EFS) provides a shared POSIX file system mountable by many instances concurrently.

What is Amazon EFS and what makes it suited to shared/resilient access?

EFS is a fully managed, elastic NFS (NFSv4) file system that can be mounted concurrently by thousands of EC2/containers/Lambda across multiple AZs in a Region. It grows/shrinks automatically and (Standard class) stores data redundantly across multiple AZs.

Compare EFS Standard storage class with EFS One Zone.

EFS Standard stores data across multiple AZs for highest availability/durability. EFS One Zone stores data in a single AZ at lower cost, suitable for non-critical or easily reproducible data. Both support lifecycle management to Infrequent Access tiers.

What is Amazon FSx and name two of its file-system flavors.

FSx provides fully managed third-party file systems. Examples: FSx for Windows File Server (SMB, Active Directory integration), FSx for Lustre (high-performance computing), FSx for NetApp ONTAP, and FSx for OpenZFS. Multi-AZ deployment options provide HA.

What is an S3 Lifecycle policy used for?

It automates moving objects between storage classes (transition actions, e.g., Standard -> Standard-IA -> Glacier) and deleting them (expiration actions) based on object age, optimizing cost over the data's lifecycle.

What is S3 Versioning and how does it support data integrity/resilience?

Versioning keeps multiple versions of an object in a bucket. It protects against accidental overwrites and deletions (a delete adds a delete marker, prior versions remain recoverable) and is a prerequisite for replication and MFA Delete.

How does S3 verify data integrity during upload and at rest?

S3 validates uploads using checksums (e.g., MD5 via Content-MD5/ETag, or stronger algorithms like SHA-256/CRC32 with the additional checksum feature). At rest it continuously verifies and self-heals data using checksums across redundant copies.

What is S3 Object Lock and which DR/compliance need does it serve?

S3 Object Lock enforces write-once-read-many (WORM) protection, preventing object versions from being deleted or overwritten for a retention period or under a legal hold. It supports compliance and protects backups against ransomware/accidental deletion (immutability).