🇺🇸 AWS Certified Solutions Architect · subject
AWS Certified Solutions Architect Design Secure Architectures Syllabus
The Design Secure Architectures syllabus for AWS Certified Solutions Architect in full: 3 chapters, 11 topics and 37 sub-topics, every one listed below, plus 70 Design Secure Architectures flashcards with answers.
Design Secure Architectures syllabus: all 3 chapters
Expand any chapter to see its topics and sub-topics.
-
Identity and Access Management (IAM)
4 topics- IAM Core Identities and Policies
- IAM users, groups, and the root account
- Identity-based vs resource-based policies
- Policy structure: Effect, Action, Resource, Condition
- Policy evaluation logic and explicit deny precedence
- Managed vs inline policies and permissions boundaries
- IAM Roles and Temporary Credentials
- Roles for EC2 instance profiles and AWS services
- AWS STS AssumeRole and temporary credentials
- Cross-account role access and trust policies
- Role chaining and session policies
- Federation and Centralized Access
- SAML 2.0 and OIDC web identity federation
- AWS IAM Identity Center (SSO) and permission sets
- Amazon Cognito user pools and identity pools
- Multi-Account Governance
- AWS Organizations and organizational units (OUs)
- Service Control Policies (SCPs)
- Consolidated billing and account isolation strategy
- IAM Core Identities and Policies
-
Data Protection and Encryption
4 topics- Encryption at Rest with AWS KMS
- AWS-managed vs customer-managed keys (CMKs)
- Key policies, grants, and rotation
- Envelope encryption and data key concepts
- S3, EBS, RDS, and DynamoDB encryption integration
- Encryption in Transit
- TLS termination at ELB and CloudFront
- AWS Certificate Manager (ACM) provisioning and renewal
- Enforcing HTTPS via policies and viewer protocol settings
- Secrets and Credential Management
- AWS Secrets Manager and automatic rotation
- Systems Manager Parameter Store SecureString
- Choosing Secrets Manager vs Parameter Store
- CloudHSM and Compliance
- CloudHSM dedicated hardware use cases
- Data residency and compliance considerations
- Encryption at Rest with AWS KMS
-
Network and Infrastructure Security
3 topics- VPC Security Controls
- Security groups (stateful) vs network ACLs (stateless)
- Subnet segmentation: public, private, isolated tiers
- VPC endpoints to keep traffic off the public internet
- Edge and Application Protection
- AWS WAF rules and managed rule groups
- AWS Shield Standard and Advanced DDoS protection
- AWS Firewall Manager for centralized policy
- Threat Detection and Monitoring
- Amazon GuardDuty threat detection
- Amazon Inspector vulnerability assessment
- AWS Security Hub aggregated findings
- Amazon Macie for sensitive data discovery
- VPC Security Controls
AWS Certified Solutions Architect Design Secure Architectures flashcards — 70 questions with answers
In AWS IAM, what are the three core identity types and how do they differ?
Users (long-term credentials for a person/app), Groups (a collection of users for shared permissions; not an identity you can authenticate as), and Roles (an identity with temporary credentials assumed by trusted principals).
What is the difference between an IAM identity-based policy and a resource-based policy?
Identity-based policies attach to users, groups, or roles and define what that principal can do. Resource-based policies attach directly to a resource (e.g., S3 bucket, SQS queue) and specify which principals may access it, including cross-account principals.
In an IAM policy, what is the effective result when an explicit Deny conflicts with an explicit Allow?
The explicit Deny always wins. IAM evaluation logic: by default deny, an explicit Allow overrides default deny, but any explicit Deny overrides any Allow.
What are the required elements that make up an IAM JSON policy statement?
Effect (Allow/Deny), Action (API operations), and Resource (ARNs). Optional elements include Principal, Condition, and Sid. Principal is required in resource-based policies.
What is the difference between an AWS managed policy, a customer managed policy, and an inline policy?
AWS managed policies are created and maintained by AWS and reusable. Customer managed policies are created by you and reusable across identities. Inline policies are embedded directly in a single identity with a strict 1:1 relationship and cannot be reused.
What is an IAM permissions boundary and what does it control?
A permissions boundary is a managed policy that sets the maximum permissions an identity-based policy can grant to a user or role. Effective permissions = intersection of the boundary and the identity-based policy; it limits but never grants permissions.
What is the IAM policy evaluation order across SCPs, permissions boundaries, identity, and resource policies?
Access is granted only if no explicit Deny exists anywhere AND the action is allowed by each applicable scope: Organizations SCPs, permissions boundary, identity-based policy (and/or resource-based policy). It is an intersection of allows with deny-override.
What is the AWS principle of least privilege and one tool AWS provides to help achieve it?
Grant only the permissions required to perform a task. IAM Access Analyzer can generate least-privilege policies based on CloudTrail access activity, and last-accessed information helps remove unused permissions.
Why should an IAM role be preferred over embedding IAM user access keys in an EC2 application?
Roles supply automatically rotated, temporary credentials via the instance metadata service, eliminating long-term secrets stored in code/config and reducing the risk of leaked, non-expiring keys.
What is the difference between authentication and authorization in AWS IAM?
Authentication verifies identity (who you are) via credentials such as passwords, access keys, or MFA. Authorization determines what an authenticated principal is permitted to do, evaluated through IAM policies.
What AWS service issues temporary security credentials and what three components do they contain?
AWS STS (Security Token Service). Temporary credentials contain an access key ID, a secret access key, and a session token, all with a limited expiration.
What STS API is used to assume a role within or across accounts, and what does it return?
sts:AssumeRole. It returns temporary credentials (access key, secret key, session token) scoped to the assumed role's permissions for the session duration.
What is an IAM role trust policy and what does it define?
The trust policy (assume-role policy) is the resource-based policy on a role that specifies which principals (accounts, services, federated users) are allowed to assume the role via sts:AssumeRole.
What is the valid session duration range for an assumed IAM role using AssumeRole?
15 minutes minimum up to a maximum of 12 hours, bounded by the role's configured maximum session duration setting.
What is an IAM instance profile and why is it needed for EC2?
An instance profile is a container for an IAM role that is attached to an EC2 instance, allowing the instance to assume the role and obtain temporary credentials automatically. EC2 cannot use a role directly without an instance profile.
What is the purpose of an external ID when a third party assumes a role in your account?
The external ID is a shared secret in the trust policy condition that prevents the 'confused deputy' problem, ensuring the third party assumes the role only on behalf of your account and not someone else's.
How do EC2 IMDSv2 sessions improve security over IMDSv1 for retrieving role credentials?
IMDSv2 requires a session-oriented, token-based request (PUT to obtain a token, then GET with the token), mitigating SSRF attacks that could otherwise read instance role credentials from the metadata endpoint.
What is an STS session policy and how does it affect permissions?
A session policy is an inline policy passed during AssumeRole that further restricts the role's permissions for that session. Effective permissions = intersection of the role's identity policy and the session policy.
What is the difference between AssumeRoleWithSAML and AssumeRoleWithWebIdentity?
AssumeRoleWithSAML exchanges a SAML 2.0 assertion (enterprise IdP like AD FS) for temporary credentials. AssumeRoleWithWebIdentity exchanges a web identity token from an OIDC provider (e.g., Google, Cognito, login.amazon.com) for temporary credentials.
What is the difference between identity federation and creating IAM users for external/enterprise users?
Federation lets users authenticate with an existing external identity provider (SAML/OIDC) and assume roles for temporary access, avoiding the management overhead and security risk of creating and maintaining individual long-term IAM users.
Show the other 50 Design Secure Architectures flashcards
What is AWS IAM Identity Center (formerly AWS SSO) used for?
It provides centralized single sign-on and permission management across multiple AWS accounts and applications, connecting to an external identity source (its own directory, Active Directory, or external IdP) and assigning permission sets to users/groups.
In IAM Identity Center, what is a permission set?
A permission set is a collection of policies that defines the level of access; it is provisioned as an IAM role in each assigned account, mapping users/groups to permissions across the organization.
What is SAML 2.0 federation and which AWS component trusts the external IdP?
SAML 2.0 federation lets enterprise users sign in with a corporate IdP. You create a SAML identity provider entity in IAM that establishes trust, and users assume a role via AssumeRoleWithSAML using the IdP's signed assertion.
What is Amazon Cognito and what are its two main components?
Cognito provides authentication for web/mobile apps. User Pools handle sign-up/sign-in and issue tokens (user directory/authentication); Identity Pools (federated identities) exchange tokens for temporary AWS credentials to access AWS resources.
What is the difference between a Cognito User Pool and a Cognito Identity Pool?
A User Pool is a user directory that authenticates users and returns JWT tokens. An Identity Pool grants temporary AWS credentials (via STS) so authenticated or guest users can access AWS services directly.
What is AWS Organizations and what is the root of its structure?
AWS Organizations centrally manages and governs multiple AWS accounts. The structure is a tree with a management (root) account at the top, organizational units (OUs) for grouping, and member accounts.
What is a Service Control Policy (SCP) and what does it actually do?
An SCP is an Organizations policy that sets the maximum available permissions (a permission guardrail) for accounts in an OU or the organization. It never grants permissions; it only filters what IAM policies in member accounts can allow.
Do SCPs affect the AWS Organizations management account?
No. SCPs do not restrict the management (root) account or any service-linked roles, which is why running workloads in the management account is discouraged.
What is AWS Control Tower and how does it relate to Organizations?
Control Tower automates setup of a secure, multi-account 'landing zone' using AWS Organizations, with pre-packaged guardrails (preventive via SCPs and detective via Config rules), centralized logging, and an account factory for standardized account provisioning.
What is the recommended multi-account strategy for isolating production from development workloads?
Use separate AWS accounts per environment/workload under one organization, applying SCP guardrails per OU. Account boundaries provide strong blast-radius isolation, separate billing, and independent security controls.
What is consolidated billing in AWS Organizations and one cost benefit it provides?
Consolidated billing aggregates charges from all member accounts into the management account's single bill. Benefit: combined usage can reach volume pricing tiers and pooled Reserved Instance / Savings Plans discounts across accounts.
What is the difference between envelope encryption and direct encryption with a KMS key?
In envelope encryption, KMS generates a data key; the data key encrypts your data locally, and KMS encrypts (wraps) the data key with the KMS key. This avoids sending large payloads to KMS (limited to 4 KB) and improves performance.
What is the difference between an AWS managed KMS key, a customer managed KMS key, and an AWS owned key?
AWS managed keys (aws/service) are created/managed by AWS per service with limited control. Customer managed keys are created and fully controlled by you (policies, rotation, deletion). AWS owned keys are shared, invisible keys AWS uses internally at no cost and you cannot manage them.
How does automatic key rotation work for a customer managed symmetric KMS key?
When enabled, KMS rotates the backing cryptographic material annually (default) while keeping the same key ID/ARN. Old material is retained to decrypt previously encrypted data; rotation is transparent to applications.
What is a KMS key policy and how does it relate to IAM policies for key access?
Every KMS key has a key policy (resource-based) that is the primary control over who can use/manage it. Access requires permission in the key policy; IAM policies alone cannot grant access unless the key policy delegates to IAM (via the account-root statement).
What is a KMS grant and when would you use it instead of a key policy?
A grant is a programmatic, temporary delegation of specific KMS permissions to a principal, often used by AWS services to encrypt/decrypt on your behalf. It's preferred for fine-grained, revocable, often short-lived access without editing the key policy.
What is the mandatory waiting period when scheduling deletion of a KMS key, and why?
7 to 30 days (default 30). The waiting period is enforced because deleting a key is irreversible and makes all data encrypted under it permanently unrecoverable, allowing time to cancel.
What is the difference between a symmetric and an asymmetric KMS key?
Symmetric keys use a single 256-bit key for both encrypt and decrypt (never leaves KMS) and are the default for most AWS encryption. Asymmetric keys use a public/private key pair for encrypt/decrypt or sign/verify, useful when the public key must be shared outside AWS.
What does SSE-S3 vs SSE-KMS vs SSE-C mean for S3 server-side encryption?
SSE-S3: S3-managed keys (AES-256), no key control. SSE-KMS: keys managed in AWS KMS with auditing and access policies. SSE-C: customer-provided keys, where you supply the key on each request and AWS does not store it.
What protocol and which TLS feature provides encryption in transit for AWS API calls and web traffic?
TLS (Transport Layer Security) over HTTPS. AWS service endpoints and load balancers use TLS certificates to encrypt data between client and service in transit.
What is AWS Certificate Manager (ACM) and a key limitation on where its public certificates can be used?
ACM provisions, manages, and auto-renews TLS/SSL certificates. ACM public certificates integrate only with supported AWS services (e.g., ELB/ALB, CloudFront, API Gateway) and cannot be exported for use on standalone EC2 instances.
What is the difference between TLS termination and TLS passthrough on a load balancer?
TLS termination decrypts traffic at the load balancer (ALB/NLB), then optionally re-encrypts to targets; it offloads crypto and enables L7 inspection. TLS passthrough (NLB) forwards encrypted traffic unmodified to the backend, which holds the certificate (end-to-end encryption).
How can you enforce that S3 only accepts requests over encrypted (HTTPS) connections?
Add a bucket policy that denies actions when the condition aws:SecureTransport is false, forcing all access to use TLS/HTTPS.
What is AWS Secrets Manager and its key feature over plain Parameter Store?
Secrets Manager securely stores secrets (DB credentials, API keys) encrypted with KMS and provides built-in automatic rotation using Lambda functions, plus fine-grained resource policies and cross-region replication.
What is the difference between AWS Secrets Manager and SSM Parameter Store SecureString for storing secrets?
Both encrypt with KMS. Secrets Manager adds native automatic rotation, generates random secrets, and cross-region replication (paid per secret). Parameter Store SecureString is lower/no cost and good for config/secrets but has no built-in rotation (rotation must be custom).
Why is storing credentials in Secrets Manager and retrieving at runtime preferred over hardcoding them in code or environment variables?
It centralizes secrets, encrypts them at rest with KMS, enables automatic rotation and auditing via CloudTrail, and prevents secrets from leaking through source control, AMIs, or container images.
What is CloudHSM and how does it differ from AWS KMS in terms of key control?
CloudHSM provides dedicated, single-tenant FIPS 140-2 Level 3 hardware security modules where you have exclusive control of keys; AWS cannot access them. KMS is multi-tenant and managed by AWS. CloudHSM suits strict compliance and custom cryptographic needs.
What FIPS 140-2 validation levels are associated with AWS KMS and AWS CloudHSM respectively?
KMS HSMs are validated to FIPS 140-2 Level 3 (and offer FIPS endpoints), and CloudHSM provides FIPS 140-2 Level 3 validated single-tenant HSMs giving the customer sole control of key material.
What is a KMS custom key store and how does it relate to CloudHSM?
A custom key store backs KMS keys with a CloudHSM cluster you control, combining the KMS API integration with AWS services and the single-tenant, customer-controlled HSM key material required for stricter compliance.
What is AWS Artifact and what does it provide for compliance?
AWS Artifact is a self-service portal for on-demand access to AWS compliance reports and documents (e.g., SOC, PCI DSS, ISO) and agreements, used as evidence for audits.
What is the difference between a security group and a network ACL in a VPC?
Security groups are stateful, operate at the instance/ENI level, support only allow rules, and evaluate all rules. Network ACLs are stateless, operate at the subnet level, support allow and deny rules, and are evaluated in numbered order.
Why must a network ACL have explicit outbound rules for return traffic while a security group does not?
NACLs are stateless, so response traffic must be explicitly allowed (often on ephemeral ports). Security groups are stateful and automatically allow return traffic for any permitted inbound/outbound flow.
What is the difference between a public subnet and a private subnet in a VPC?
A public subnet has a route to an internet gateway, allowing direct inbound/outbound internet access. A private subnet has no direct internet route; outbound internet access requires a NAT gateway/instance, providing isolation for backend resources.
What is a VPC endpoint, and how do Gateway vs Interface endpoints differ?
A VPC endpoint privately connects a VPC to AWS services without traversing the internet. Gateway endpoints (route-table entries) support only S3 and DynamoDB. Interface endpoints (PrivateLink ENIs with private IPs) support most other services.
What is the security benefit of a NAT gateway for private subnet instances?
It allows instances in private subnets to initiate outbound internet connections (e.g., updates) while preventing any unsolicited inbound connections from the internet, preserving isolation.
What is AWS Network Firewall and how does it differ from security groups/NACLs?
Network Firewall is a managed, stateful network/VPC-level firewall with IDS/IPS, domain filtering, and deep packet inspection using Suricata-compatible rules. It provides far richer controls than the basic allow/deny of security groups and NACLs.
What is AWS WAF and at which layer does it protect applications?
AWS WAF is a web application firewall operating at Layer 7 (HTTP/HTTPS). It filters requests by rules (SQL injection, XSS, rate limiting, geo, IP) and attaches to CloudFront, ALB, API Gateway, or AppSync.
What is AWS Shield Standard versus AWS Shield Advanced?
Shield Standard is free, automatic protection against common L3/L4 DDoS attacks. Shield Advanced is paid and adds enhanced detection, protection for higher-layer attacks, 24/7 DDoS Response Team access, WAF integration, and DDoS cost protection (billing reimbursement).
How does Amazon CloudFront improve security at the edge for an application?
CloudFront serves content from edge locations over TLS, integrates with AWS WAF and Shield for DDoS/L7 protection, supports Origin Access Control to restrict S3 origins, and can use signed URLs/cookies and geo-restriction to control access.
What is Origin Access Control (OAC) for CloudFront with S3?
OAC restricts an S3 bucket so it can only be accessed through the CloudFront distribution (not directly), by having CloudFront sign origin requests; the bucket policy denies all other access. It replaces the legacy Origin Access Identity (OAI).
What is Amazon GuardDuty and what data sources does it analyze?
GuardDuty is a managed threat-detection service using ML and threat intelligence to find malicious or unauthorized activity. It analyzes CloudTrail events, VPC Flow Logs, and DNS logs (plus optional S3, EKS, malware, and RDS protection) without deploying agents.
What is AWS CloudTrail and what does it record?
CloudTrail records account activity by logging API calls and management/data events (who, what, when, from where) for governance, compliance, auditing, and incident investigation. Logs can be delivered to S3 and CloudWatch Logs.
What is the difference between AWS CloudTrail and Amazon CloudWatch?
CloudTrail audits API activity (who did what to which resource) for governance/security. CloudWatch monitors operational performance via metrics, logs, alarms, and dashboards. CloudTrail = audit trail; CloudWatch = monitoring/observability.
What is AWS Config and what core question does it answer?
AWS Config records and evaluates resource configurations over time, answering 'what is my configuration and is it compliant?'. It tracks configuration history/changes and uses Config rules to flag non-compliant resources.
What is Amazon Inspector and what does it assess?
Amazon Inspector is an automated vulnerability management service that continuously scans EC2 instances, container images in ECR, and Lambda functions for software vulnerabilities (CVEs) and unintended network exposure.
What is Amazon Macie and what is its primary purpose?
Macie is a managed data security service that uses machine learning and pattern matching to discover, classify, and protect sensitive data (e.g., PII, credentials) stored in Amazon S3.
What is AWS Security Hub and how does it relate to GuardDuty, Inspector, and Macie?
Security Hub is a central dashboard that aggregates, normalizes, and prioritizes security findings from GuardDuty, Inspector, Macie, and partner tools, and runs automated compliance checks against standards (CIS, PCI DSS, AWS Foundational Best Practices).
What are VPC Flow Logs and what security purpose do they serve?
VPC Flow Logs capture metadata about IP traffic to/from network interfaces (source/dest IPs, ports, protocol, accept/reject). They support troubleshooting, traffic analysis, and detecting anomalous or unauthorized network activity, and feed GuardDuty.
What is AWS IAM Access Analyzer and what does it detect?
Access Analyzer uses automated reasoning to identify resources (S3 buckets, roles, KMS keys, etc.) shared with external entities or made public, flagging unintended access, and it can validate and generate least-privilege policies.
Under the AWS Shared Responsibility Model, who is responsible for encryption configuration, IAM policies, and OS patching on EC2?
The customer. AWS is responsible for security OF the cloud (hardware, global infrastructure, managed service software); the customer is responsible for security IN the cloud, including IAM, encryption settings, network/firewall config, and guest OS patching on EC2.
AWS Certified Solutions Architect Design Secure Architectures FAQ
What is in the AWS Certified Solutions Architect Design Secure Architectures syllabus?
AWS Certified Solutions Architect Design Secure Architectures is split into 3 chapters: Identity and Access Management (IAM), Data Protection and Encryption and Network and Infrastructure Security. Together they contain 11 topics and 37 sub-topics.
How many chapters are in AWS Certified Solutions Architect Design Secure Architectures?
3 chapters, containing 11 topics and 37 sub-topics. The largest chapter is Identity and Access Management (IAM), with 4 topics.
Are there flashcards for AWS Certified Solutions Architect Design Secure Architectures?
Yes — a 70-card Design Secure Architectures deck, printed in full in the flashcards section of this page and available in the Examius app with spaced repetition. The free plan covers one exam of your choice, including its flashcards; Examius Pro, a one-time purchase, unlocks every other exam and deck.