AWS Network Threat Modeling & Mitigation Strategies Study Guide
Developing a threat model and identifying appropriate mitigation strategies for a given network architecture
AWS Network Threat Modeling & Mitigation Strategies
This guide explores the structured approach to developing threat models and implementing robust mitigation strategies within AWS network architectures, specifically tailored for the ANS-C01 exam.
Learning Objectives
After studying this guide, you should be able to:
- Define the steps of a comprehensive threat modeling process.
- Identify assets, threats, and vulnerabilities within a specific network architecture.
- Map appropriate AWS security services to specific threat categories.
- Architect secure patterns such as Three-Tier, Perimeter VPC, and Hub-and-Spoke.
- Implement compliance testing and ongoing monitoring to validate security controls.
Key Terms & Glossary
- Threat Modeling: A structured process to identify, quantify, and address security risks associated with a system.
- DMZ (Demilitarized Zone): Also known as an untrusted network architecture; a sub-network that exposes an organization's external-facing services to an untrusted network (usually the Internet).
- Perimeter VPC: A dedicated VPC that acts as a gatekeeper for all traffic entering or leaving an AWS environment, often hosting inspection appliances.
- Blast Radius: The potential extent of damage that can be caused by a single security incident or technical failure.
- Shared Responsibility Model: The AWS security framework where AWS manages security "of" the cloud, and the customer manages security "in" the cloud.
The "Big Idea"
Security in AWS is not a static state but a continuous lifecycle. Threat modeling is the foundation of this lifecycle; you cannot protect what you do not understand. By systematically identifying assets and potential attackers, you move from a "hope-based" security posture to a "risk-based" architecture where controls (like NACLs, Security Groups, and WAFs) are applied precisely where they are most effective.
Formula / Concept Box
| Threat Modeling Step | Key Action | AWS Tooling/Artifacts |
|---|---|---|
| 1. Identify Assets | Catalog servers, data, and microservices. | AWS Config, Tag Editor |
| 2. Identify Risks | Find unpatched software or misconfigurations. | Amazon Inspector, Trusted Advisor |
| 3. Assess Impact | Determine likelihood and severity of breaches. | Risk Matrix |
| 4. Identify Actors | Research internal vs. external threat agents. | Threat Intel / GuardDuty |
| 5. Mitigation | Apply controls (Encryption, Segmentation). | KMS, Security Groups, IAM |
Hierarchical Outline
-
The Threat Modeling Lifecycle
- Inventory Phase: Identifying what needs protection (data, compute, IP).
- Analysis Phase: Investigating threat actors (nation-states, hackers, insiders).
- Prioritization: Assessing the likelihood vs. impact of each threat.
-
Standard Secure Architectures
- Three-Tier Architecture: Separation of Web, App, and Database into distinct subnets.
- Hub-and-Spoke: Centralized control via AWS Transit Gateway connecting multiple spoke VPCs.
- Perimeter VPC (DMZ): Isolating public-facing resources to protect the internal network.
-
Core Mitigation Mechanisms
- Inbound Protection: AWS WAF (Layer 7), AWS Shield (DDoS), AWS Network Firewall.
- Inter-VPC Security: Security Groups (stateful), NACLs (stateless), VPC Endpoint Policies.
- Data Protection: Encryption at rest (KMS) and in transit (TLS/IPsec).
-
Validation & Compliance
- Automated Testing: Using AWS Config and Lambda for drift detection.
- Active Testing: Penetration testing and vulnerability scanning with Amazon Inspector.
Visual Anchors
The Threat Modeling Workflow
Standard Three-Tier Segmentation
\begin{tikzpicture}[node distance=1.5cm, every node/.style={rectangle, draw, minimum width=3cm, minimum height=1cm, align=center}]
% Nodes \node (internet) [fill=gray!20] {Public Internet}; \node (igw) [below of=internet] {Internet Gateway}; \node (web) [below of=igw, fill=blue!10] {Web Tier Subnet$Public)}; \node (app) [below of=web, fill=green!10] {Application Tier Subnet$Private)}; \node (db) [below of=app, fill=red!10] {Database Tier Subnet$Private)};
% Connections \draw [->, thick] (internet) -- (igw); \draw [->, thick] (igw) -- (web); \draw [->, thick] (web) -- node[right, draw=none] {SG / NACL} (app); \draw [->, thick] (app) -- node[right, draw=none] {SG / NACL} (db);
% Legend/Annotation \node [right of=web, xshift=3cm, draw=none, text width=4cm] {\small \textbf{Inbound}: WAF/Shield}; \node [right of=db, xshift=3cm, draw=none, text width=4cm] {\small \textbf{Storage}: Encryption at Rest};
\end{tikzpicture}
Definition-Example Pairs
- Segmentation: The act of splitting a network into smaller parts to improve performance and security.
- Example: Placing a database in a private subnet with no route to the Internet Gateway, allowing access only from the application tier.
- Mitigation Strategy: A specific action or tool used to reduce the risk associated with a threat.
- Example: Implementing AWS Shield Advanced to mitigate the risk of a high-volume Layer 3/4 DDoS attack.
- VPC Endpoint Policy: An IAM resource policy attached to a VPC endpoint to control which principals can use the endpoint to access a service.
- Example: Creating a policy for an S3 Gateway Endpoint that only allows access to a specific corporate bucket, preventing data exfiltration to personal accounts.
Worked Examples
Scenario: Securing a Multi-Region Web Application
Goal: Protect a web application from SQL injection and ensure cross-region traffic is encrypted.
- Threat Model Discovery:
- Asset: Customer Database (RDS).
- Threat: SQL Injection via the web tier.
- Vulnerability: Exposed public endpoints.
- Mitigation Plan:
- Layer 7: Deploy AWS WAF on the Application Load Balancer (ALB) with SQLi injection protection rules.
- Layer 4: Configure Security Groups to allow only traffic from the ALB to the App instances, and only from App instances to the DB.
- Encryption: Enable TLS 1.2 for all data in transit. Use AWS KMS to encrypt the RDS volume at rest.
- Validation:
- Run Amazon Inspector to check for EC2 vulnerabilities.
- Use AWS Config to ensure no security groups allow
0.0.0.0/0on port 3306 (MySQL).
Checkpoint Questions
- What is the primary difference between a Security Group and a Network ACL in terms of statefulness?
- Why should a Perimeter VPC be isolated from other VPCs in the environment?
- Which AWS service is best suited for automating the identification of unpatched software on EC2 instances?
- Describe the function of a "Hub" in a Hub-and-Spoke architecture.
▶Click to see answers
- Security Groups are stateful (responses are allowed automatically); NACLs are stateless (rules must be defined for both inbound and outbound traffic).
- To ensure all traffic is forced through a centralized point of inspection before reaching internal resources, minimizing the risk of lateral movement.
- Amazon Inspector.
- It acts as a central point of control, usually containing a Transit Gateway or centralized firewall, to manage traffic between spokes and external connections.
Muddy Points & Cross-Refs
- NACL vs. Security Group: Students often struggle with where to apply each. Remember: NACLs are your subnet-level "fences" (broad), while Security Groups are your instance-level "doors" (specific).
- Perimeter VPC vs. Transit Gateway: A Perimeter VPC is an architectural pattern, while Transit Gateway is a service used to implement patterns like Hub-and-Spoke. You often use them together.
- Compliance Testing: Don't confuse AWS Config (resource configuration history) with AWS CloudTrail (API call history).
Comparison Tables
Inbound vs. Outbound Traffic Controls
| Control Point | Inbound Focus | Outbound Focus |
|---|---|---|
| AWS WAF | SQLi, XSS, Bot Control | N/A |
| Network Firewall | Deep Packet Inspection (DPI) | Domain/URL Filtering |
| Gateway Load Balancer | Third-party IPS/IDS Appliances | Centralized Egress Inspection |
| Security Groups | Limiting access to App ports | Preventing data exfiltration |
[!IMPORTANT] Threat models must be reviewed on a regular basis. As your network architecture or the global threat landscape evolves, your security controls must be updated to remain effective.