CSA (312-39) SOC Simulation Lab
Scenario Context
You are a Senior SOC Analyst reviewing the onboarding architecture for a new fleet of Linux web servers at MedSecure Cloud. A junior engineer has drafted the ingestion pipeline using Elastic Beats (SIEM Agents) and is confused about which component handles data normalization, correlation, and visualization. They have provided their filebeat.yml configuration for your review to understand what happens at the edge before data reaches the central engine.
Security Environment
Artifact: /etc/filebeat/filebeat.yml (SIEM Agent Configuration)
Question
1. Collecting data received from various devices sending data to SIEM before forwarding it to the central engine.
2. Normalizing data received from various devices sending data to SIEM before forwarding it to the central engine.
3. Co-relating data received from various devices sending data to SIEM before forwarding it to the central engine.
4. Visualizing data received from various devices sending data to SIEM before forwarding it to the central engine.
Expert Insight
1. What is happening
The SOC is designing an ingestion pipeline. We deploy SIEM Agents (like Splunk Universal Forwarders, Winlogbeat, or Filebeat) to edge devices. Their primary job is to grab logs from local files or Event Tracing frameworks and ship them over the network to a central log collector or indexer.
2. The True Role of SIEM Agents
In standard architecture, agents handle:
- Collection (1): Tailing log files, hooking into APIs, and gathering raw telemetry.
- Normalization (2): Standardizing formats (e.g., converting dates, mapping fields to a Common Information Model like ECS, or dropping noisy fields as seen in the configuration snippet above). *Note: Some "lightweight" agents only collect, pushing normalization to an intermediate tier like Logstash or Heavy Forwarders.*
3. Why Correlation and Visualization belong to the Central Engine
Correlation (3): Requires evaluating logs from multiple different devices simultaneously to spot a complex attack pattern. An agent on a single endpoint lacks the global visibility required to correlate.
Visualization (4): Generating dashboards, charts, and SOC operational screens requires a web server and querying engine, which resides at the top tier of the SIEM architecture, far removed from the edge agent.
4. Real-world SOC application
When tuning SIEM performance, SOC engineers must decide where to normalize data. "Heavy" agents consume more CPU/RAM on the host server because they parse and normalize data locally before sending. "Universal" or lightweight agents simply collect and forward raw data, offloading the CPU-intensive normalization work to central indexers. Understanding this balance is critical to prevent your security tools from crashing production web servers.
MINI LESSON: The 4 Tiers of SIEM Architecture
- Tier 1: Collection (Agents/Forwarders) - Gathers logs from endpoints, cloud trails, and network appliances. Optionally parses and filters.
- Tier 2: Ingestion & Indexing - Receives data, completes normalization, indexes data for fast searching, and writes to disk.
- Tier 3: Correlation Engine - Runs scheduled searches and rule analytics across the indexed data to generate Alerts.
- Tier 4: Presentation / Visualization - The UI dashboard where SOC analysts run threat hunts, view charts, and manage cases.
Enhance your threat detection capabilities and master the CSA curriculum.
Explore more CSA simulations