SAP-C02 Cheat Sheet and Decision Matrices · Baitaphish

title: "SAP-C02 Decision Cheat Sheet" summary: "A compact framework for evaluating complex AWS architecture, migration, modernization, operations, and governance tradeoffs."

Professional architecture sequence

Identify business outcome, owner, users, data, and authoritative systems.
Extract every constraint: time, scale, consistency, RTO/RPO, security, compliance, operations, migration, and cost.
Draw identity, request, network, data, dependency, deployment, and evidence paths.
Compare at least two viable patterns.
Reject alternatives using a stated constraint, not preference.
Test failure, quota, scaling, change, rollback, recovery, and cost behavior.
State who operates and pays for the design.

Keep emerging pretest subjects—generative-AI guardrails, AgentCore identity, and human oversight—separate from the four scored domain weights. They may appear as unscored questions according to the current guide and must not displace core architectural preparation.

Use the matrices from the requirement column. Change one constraint and explain whether the answer changes; that is closer to professional-level reasoning than memorizing service pairs.

Organization and account boundary

Use accounts as ownership, isolation, and quota boundaries—not merely folders.

Organization and account boundary: Use accounts as ownership, isolation, and quota boundaries—not merely folders.
Need	Primary mechanism	Tradeoff
Independent security boundary	Separate account	More governance and shared-service design
Policy inheritance	OU and organization policies	OU churn can complicate exceptions
Shared capability	Dedicated shared-services account	Must control blast radius and dependencies
Workforce access	IAM Identity Center/federation	Requires lifecycle and permission-set governance

Objectives: 1.2, 1.4, 1.5

I can explain this matrix without notes

Connectivity selection

Choose connectivity from bandwidth, latency, reach, encryption, convergence, and operations.

Connectivity selection: Choose connectivity from bandwidth, latency, reach, encryption, convergence, and operations.
Requirement	Pattern	Watch
Quick encrypted hybrid	Site-to-Site VPN	Internet path and throughput
Consistent private capacity	Direct Connect with resilient design	Lead time and encryption requirement
Many VPCs/accounts	Transit Gateway or Cloud WAN	Route domains, cost, inspection
Service-specific private access	PrivateLink	Provider/consumer DNS and policy

Objectives: 1.1

I can explain this matrix without notes

Integration pattern

Match delivery semantics and failure isolation to the business process.

Integration pattern: Match delivery semantics and failure isolation to the business process.
Need	Pattern	Critical control
Decouple one consumer	Queue	Visibility, retry, DLQ, idempotency
Fan-out events	Topic/event bus	Filtering, schema, target failure
Ordered stream/replay	Stream	Partition key, retention, consumer lag
Stateful orchestration	Workflow	Timeout, compensation, idempotency

Objectives: 2.1, 2.4

I can explain this matrix without notes

Disaster-recovery strategy

RTO/RPO and operating ability determine the pattern.

Disaster-recovery strategy: RTO/RPO and operating ability determine the pattern.
Pattern	RTO/RPO tendency	Operational cost
Backup/restore	Longest	Lowest steady state
Pilot light	Long to moderate	Core data/services running
Warm standby	Moderate to short	Scaled-down complete environment
Active/active	Shortest when correctly designed	Highest consistency and operations complexity

Objectives: 1.3, 2.2, 2.4, 3.4

I can explain this matrix without notes

Data service selection

Start from access patterns, consistency, scale, recovery, and ownership.

Data service selection: Start from access patterns, consistency, scale, recovery, and ownership.
Workload	Candidate family	Decision signal
Relational transactions	Aurora/RDS	SQL, constraints, joins, transaction model
Key-value at scale	DynamoDB	Known access patterns and partition design
Object/data lake	S3 and analytical services	Durability, lifecycle, open formats, query separation
Analytics warehouse	Redshift	Columnar analytical workload and ecosystem

Objectives: 2.5, 2.6

I can explain this matrix without notes

Migration strategy

Use business value, constraints, and dependency evidence.

Migration strategy: Use business value, constraints, and dependency evidence.
Strategy	Use when	Risk
Retain/retire	No current migration value	Hidden dependency or unsupported risk
Rehost	Speed and low change dominate	Carries technical debt
Replatform	Managed improvement with limited change	Compatibility and behavior differences
Refactor	Business value justifies architectural change	Time, skill, scope, migration risk

Objectives: 4.1, 4.2, 4.3, 4.4

I can explain this matrix without notes

Improvement prioritization

Use evidence, reversibility, and business impact.

Improvement prioritization: Use evidence, reversibility, and business impact.
Signal	Decision	Proof
Frequent incident	Reduce failure cause/blast radius	Incident rate, recovery time, test
Performance saturation	Change constrained layer	Representative load and telemetry
Unowned cost	Establish attribution before optimization	Unit cost and owner
Manual risky change	Automate with guardrails	Failure rate, rollback, audit

Objectives: 3.1, 3.2, 3.3, 3.4, 3.5

I can explain this matrix without notes