AWS SAP · Domain 2 of 4

AWS SAP Domain 2: Design for New Solutions

This is the largest SAP-C02 domain. It covers designing greenfield architectures from requirements: deployment strategies, business continuity, security, reliability and performance, justifying every choice against competing constraints.

Exam weight
29%
Official task statements
5
Estimated questions estimate
~19
Suggested study time estimate
40–55 hrs

About 11 min read Reviewed October 2026

Overview

Design for New Solutions is the heaviest domain and the one closest to what people imagine an architect exam to be: a greenfield requirement arrives and you design for it. It has five task statements covering deployment strategy, business continuity, security controls, reliability and performance. You translate business and technical requirements, such as performance targets, recovery objectives, compliance constraints and budget ceilings, into an architecture and justify each choice. It spans deployment and rollback mechanisms, disaster-recovery strategies matched to RTO and RPO, security controls built in from the start, resilient and scalable design, and performance patterns such as caching and the right data store.

At 29%, this domain carries the most questions. Its scenarios present competing constraints where several designs work, and reward the one that best balances them. The exam tests judgment across all Well-Architected pillars at once, which is what earns the professional label.

How it's tested

SAP-C02 questions in this domain are greenfield design scenarios with multiple requirements. Read every constraint, since several options may technically work and the exam wants the one that best satisfies all of them. Reward the design that balances all the constraints rather than maximizing one pillar.

  • Match the DR strategy

    RTO and RPO and a budget are given, and you choose backup and restore, pilot light, warm standby or active-active.

  • Design the deployment

    A release requirement is described, and you choose a deployment and rollback strategy such as blue/green or canary.

  • Balance the pillars

    Competing performance, cost, security and reliability constraints are given, and you choose the design that best satisfies all of them.

Reasoning it through

A new application must recover within minutes of a Regional failure but the budget rules out running a full duplicate environment at all times. Active-active meets the recovery time but costs the most, and backup and restore is cheap but far too slow. A warm standby keeps a scaled-down copy running in a second Region that scales up on failover, meeting the recovery objective within budget, which is the balanced answer the exam expects.

Task statements 5

Official task statements from the Amazon Web Services outline, in outline order.

2.1 Design a deployment strategy to meet business requirementsDeployment and rollback mechanisms and automation.
What it means
This covers selecting services and strategies to deploy workloads with safe rollback: blue/green and canary deployments, infrastructure as code with CloudFormation and the CDK, CI/CD pipelines, and delegating deployment complexity to managed services.
What to understand
Deployment strategy balances speed against risk and must include rollback. Blue/green switches between environments for fast rollback, while canary shifts traffic gradually. Infrastructure as code makes deployments repeatable. The right choice depends on the risk tolerance and release cadence the scenario states.
How it shows up
A deployment requirement is described, and you must choose the strategy, rollback mechanism and automation.
Common confusion
Choosing an all-at-once deployment where the requirement implies controlled rollout and rollback.
2.2 Design a solution to ensure business continuityDisaster recovery strategies matched to RTO and RPO.
What it means
This covers disaster-recovery strategies, backup and restore, pilot light, warm standby and active-active multi-site, matched to recovery time and recovery point objectives, cross-Region replication, failover with Route 53, and data backup across accounts and Regions.
What to understand
The DR strategy is driven by RTO and RPO and bounded by budget. Lower objectives cost more: active-active is near-instant and most expensive, backup and restore is cheapest and slowest. Warm standby and pilot light sit between. The exam always ties the choice to the stated objectives and cost.
How it shows up
RTO, RPO and budget are given, and you must choose the DR strategy that fits.
Common confusion
Over-provisioning active-active for a loose RTO, or under-provisioning backup and restore for a tight one.
2.3 Determine security controls based on requirementsBuilding security in from the start across the design.
What it means
This covers designing security into a new solution: identity and access with IAM and federation, encryption with KMS, network security controls, secrets management, data protection and compliance controls, and threat protection with WAF, Shield and GuardDuty.
What to understand
Security is designed in, not added later. Identity, encryption, network isolation and compliance controls are chosen to meet the requirements from the start. The right control depends on the data sensitivity, compliance regime and threat model the scenario describes.
How it shows up
A security or compliance requirement is described, and you must choose the controls to build into the design.
Common confusion
Treating security as an add-on rather than designing it into the architecture from the requirements.
2.4 Design a strategy to meet reliability requirementsResilient, scalable design with caching, buffering and replicas.
What it means
This covers designing for reliability: multi-AZ and multi-Region resilience, Auto Scaling, decoupling with queues, caching and buffering, replica design patterns, and removing single points of failure to meet availability targets.
What to understand
Reliability design removes single points of failure and absorbs load through decoupling, scaling and caching. Buffering with queues smooths spikes, and replicas scale reads and provide failover. The design must meet the stated availability target without unnecessary cost.
How it shows up
A reliability or availability requirement is described, and you must choose the resilient, scalable design.
Common confusion
Confusing high availability with fault tolerance, or adding resilience beyond what the requirement needs.
2.5 Design a solution to meet performance objectivesPerformance patterns, caching and the right service selection.
What it means
This covers meeting performance targets: selecting high-performing compute, storage and database services, caching with ElastiCache, DAX and CloudFront, content delivery, and designing for latency and throughput at scale.
What to understand
Performance design matches services and patterns to the workload: caching cuts latency and load, the right database and storage type meets throughput, and content delivery serves users close to them. The design must hit the performance target within the other constraints.
How it shows up
A performance objective is described, and you must choose the services and patterns that meet it.
Common confusion
Scaling infrastructure where caching or the right service choice would meet the performance target more efficiently.

High-yield concepts

The terms, rules and formulas this domain turns into questions most often.

  • DR strategies by RTO and RPO

    Backup and restore, pilot light, warm standby, active-active.

    Remember
    Lower objectives cost more.
    Tested as
    Business continuity questions.
    Trap
    Mismatching the strategy to the objectives.

  • Blue/green vs. canary deployment

    Switch environments versus shift traffic gradually.

    Remember
    Both enable safe rollback.
    Tested as
    Deployment questions.
    Trap
    Using all-at-once where controlled rollout is needed.

  • Warm standby

    A scaled-down copy that scales up on failover.

    Remember
    Balances cost and recovery speed.
    Tested as
    DR questions.
    Trap
    Confusing it with pilot light or active-active.

  • Route 53 failover

    DNS health checks and failover routing.

    Remember
    Redirects traffic on failure.
    Tested as
    Continuity questions.
    Trap
    Ignoring DNS TTL effects.

  • Decoupling and buffering

    Queues smooth spikes and decouple components.

    Remember
    They absorb load and add resilience.
    Tested as
    Reliability questions.
    Trap
    Tightly coupling components.

  • Caching patterns

    ElastiCache, DAX and CloudFront reduce latency and load.

    Remember
    Cache hot data and content.
    Tested as
    Performance questions.
    Trap
    Scaling instead of caching.

  • Security by design

    Identity, encryption and network controls built in from the start.

    Remember
    Design security into the architecture.
    Tested as
    Security questions.
    Trap
    Bolting on security later.

  • KMS and encryption

    Managed keys with customer control where needed.

    Remember
    Encrypt at rest and in transit.
    Tested as
    Security questions.
    Trap
    Overlooking key control requirements.

  • Multi-AZ vs. multi-Region

    Zone resilience versus Regional resilience.

    Remember
    Multi-Region handles Regional events.
    Tested as
    Reliability questions.
    Trap
    Using multi-AZ for a Regional requirement.

  • Well-Architected trade-offs

    Balancing operational excellence, security, reliability, performance, cost and sustainability.

    Remember
    The exam tests all pillars at once.
    Tested as
    Design questions.
    Trap
    Optimizing one pillar in isolation.

Side by side

Concepts candidates mix up, compared on the point that decides the answer.

DR strategies by RTO and RPO

ConceptMeaningKey differenceCommon trap
Backup and restore Restore from backups Cheapest, slowest Using it for a tight RTO
Pilot light / warm standby Minimal or scaled-down standby Balances cost and speed Confusing the two scales
Active-active Full running in two Regions Near-zero RTO, costliest Over-provisioning for loose RTO

Blue/green vs. canary vs. all-at-once

ConceptMeaningKey differenceCommon trap
Blue/green Switch between environments Fast rollback Confusing with canary
Canary Shift a small percentage first Limits blast radius Confusing with all-at-once
All-at-once All traffic immediately Fastest, riskiest Using it for risky releases

Scale vs. cache vs. decouple

ConceptMeaningKey differenceCommon trap
Scale Add capacity Handles more load Using it where caching fits
Cache Reuse results Cuts latency and load Scaling instead of caching
Decouple Queues and buffering Absorbs spikes, adds resilience Tight coupling

Exam traps

Each trap is an answer that looks right. Learn the clue that should change your mind.

  1. Active-active for loose RTO

    What candidates pick
    Design active-active multi-Region for a generous RTO.
    Why it looks right
    It recovers fastest.
    Why it's wrong
    A generous RTO can be met more cheaply with warm standby or backup and restore; active-active over-provisions.

    Clue to watch for The RTO is generous and budget is constrained.

  2. Backup-restore for tight RTO

    What candidates pick
    Use backup and restore for a near-instant recovery requirement.
    Why it looks right
    It is the cheapest option.
    Why it's wrong
    Backup and restore is too slow for a tight RTO; warm standby or active-active is needed.

    Clue to watch for The RTO is minutes or seconds.

  3. Maximize one pillar

    What candidates pick
    Choose the most resilient design regardless of the budget ceiling.
    Why it looks right
    More resilience seems better.
    Why it's wrong
    The exam rewards balancing all constraints, including cost, not maximizing one pillar.

    Clue to watch for A budget or cost constraint is stated alongside reliability.

  4. All-at-once release

    What candidates pick
    Deploy a new critical workload to all traffic at once.
    Why it looks right
    It is the fastest release.
    Why it's wrong
    A controlled strategy such as blue/green or canary with rollback fits a critical release.

    Clue to watch for The release must minimize risk and allow rollback.

  5. Scale for repeated reads

    What candidates pick
    Scale the database to handle repeated identical reads.
    Why it looks right
    More capacity handles more load.
    Why it's wrong
    Caching with ElastiCache, DAX or CloudFront meets the performance target more efficiently.

    Clue to watch for The workload is read-heavy with repeated queries.

  6. Security added later

    What candidates pick
    Design the architecture first and plan to add security afterward.
    Why it looks right
    It speeds the initial design.
    Why it's wrong
    Security must be designed in from the requirements, covering identity, encryption and network controls.

    Clue to watch for The requirements include security or compliance constraints.

Answer framework

For any Domain 2 scenario, use these steps to find the answer SAP-C02 expects.

  1. List every requirement, performance target, RTO and RPO, compliance and budget, before designing, since the best answer satisfies all of them.
  2. For business continuity, match the DR strategy to the RTO and RPO within the budget, from backup and restore to active-active.
  3. For deployment, choose a strategy with rollback that fits the risk tolerance, such as blue/green or canary.
  4. For reliability and performance, remove single points of failure and apply caching, buffering and the right service, without over-provisioning.
  5. Design security in from the start, and justify the overall design as the best balance of all Well-Architected pillars. Confirm the chosen approach beats the other workable options on the stated constraints before committing to it.

Practice question

A new application must recover within a few minutes of a full Regional outage, but the budget does not allow running a complete duplicate production environment continuously. Which disaster-recovery strategy BEST meets both requirements?

Want more like this? The AWS SAP practice test has questions for every domain.

Checklist

0 of 10

Tick each one you can do without notes. Your ticks stay in this browser.

Study strategy

This is the largest domain, so give it the most time, and practice designing against competing constraints rather than learning services in isolation. Start with business continuity, since DR strategy matched to RTO, RPO and budget is a recurring, high-value pattern: learn the four strategies cold and when each fits.

Next, learn deployment strategies with rollback, security-by-design across identity, encryption and network, reliability patterns such as decoupling and caching, and performance service selection. The unifying skill is balancing all the Well-Architected pillars at once, so whenever you pick a design, ask what it costs the other pillars.

Drill long scenario questions and force yourself to justify why the chosen design beats the others that also work. The Well-Architected and disaster-recovery whitepapers map directly here. Afterwards, move to Domain 3, continuous improvement for existing solutions. For each missed new-solution question, write down which requirement, such as RPO, latency or data residency, eliminated your choice, and drill that requirement next.

FAQ

How much of SAP-C02 is Domain 2?

Design for New Solutions is 29% of the scored content, the largest domain.

How do I choose a disaster-recovery strategy?

Match it to the required RTO and RPO within the budget. Backup and restore is cheapest and slowest, active-active is fastest and costliest, and pilot light and warm standby sit between.

What is warm standby?

A disaster-recovery strategy that keeps a scaled-down but running copy of the environment in another Region, which scales up on failover, balancing cost against recovery speed.

Why does the exam feel like it has several right answers?

Because at the professional level several designs usually work. The exam rewards the one that best balances all the stated constraints, not just one requirement.

How should security be handled in a new design?

Designed in from the requirements, covering identity, encryption, network controls and compliance, rather than added after the architecture is built.

When should I cache instead of scale?

When the workload has repeated reads or content that can be served from a cache such as ElastiCache, DAX or CloudFront, which meets performance targets more efficiently than scaling the backend.

How does Domain 2 connect to the rest of the exam?

It designs new workloads within the enterprise structure of Domain 1, which are then continuously improved in Domain 3 and may be the target of migrations in Domain 4.

Other Cloud Computing certifications

Where the same subject shows up in other exams. Cards marked "Same subject" open the matching domain.

Test your Domain 2 knowledge

You have covered deployment, business continuity, security, reliability and performance design. Test whether you can choose the design that best balances competing constraints.

Take the AWS SAP practice test Check your AWS SAP readiness

Keep going