Skip to Content
CyberDoyen SIEM 2.0 is released 🎉
Architecture

CyberDoyen SIEM Architecture


CyberDoyen SIEM is purpose-built for modern security teams who require powerful, flexible, and high-performance security monitoring. It is designed to scale seamlessly with your infrastructure, offering real-time insights and advanced threat detection without sacrificing speed or usability.


Core Architecture Overview

CyberDoyen SIEM is based on a modular, distributed architecture that ensures scalability, reliability, and performance even at enterprise scale.

ComponentDescription
Ingestion LayerCollects logs, metrics, and security events from various sources.
Processing LayerNormalizes, enriches, and correlates incoming data streams.
Storage LayerIndexes and stores structured and unstructured data efficiently.
Analytics & DetectionApplies machine learning models and detection rules for real-time threat identification.
Visualization LayerProvides intuitive dashboards, visualizations, and alerting mechanisms.
Orchestration LayerManages system configurations, plugin integrations, and data pipelines.

Architectural Diagram

Diagram Coming Soon
(Add your architecture diagram image here, e.g., /images/architecture-diagram.png)


Key Capabilities

Analytics - Real-time Processing

FeatureDescription
Advanced Threat DetectionMachine learning-powered anomaly detection identifies threats 40% faster than traditional methods.
Customizable DashboardsTailor dashboards for different teams and use cases with drag-and-drop simplicity.
High-Speed Query EngineOptimized for sub-second queries even at massive data volumes.

Scalability - Enterprise Ready

FeatureDescription
Distributed ProcessingDynamically scalable to handle event loads ranging from 1GB to 1PB+ per day.
Plugin-Based ArchitectureSupports over 150+ log sources out of the box, with the ability to add new ones quickly.
Horizontal ScalingAdd more ingestion, processing, or storage nodes as needed without downtime.

Modular Components

Ingestion Layer

  • Uses Filebeat, Syslog, API Collectors, and Custom Agents to pull logs from devices, cloud sources, endpoints, and applications.
  • Supports batch and streaming ingestion modes.

Processing Layer

  • Normalization: Converts logs into a consistent schema.
  • Enrichment: Adds metadata such as geolocation, asset tagging, threat intelligence feeds.
  • Correlation: Links related events across disparate data sources.

Storage Layer

  • Powered by Elasticsearch with custom tuning for security events.
  • Supports hot-warm-cold storage tiering for efficient long-term retention.

Analytics & Detection Engine

  • Integrated ML Models for anomaly detection, behavior profiling, and predictive analytics.
  • Rule Engine allows custom threat detection logic using simple DSL (Domain-Specific Language).

Visualization & Management

  • CyberDoyen Web UI offers:
    • Dashboards
    • Query Builder
    • Alerts and Notification System
    • User Access Controls

Data Flow

  1. Event Collection: Devices, applications, and cloud services generate events.
  2. Ingestion: Filebeat and agents forward events to ingestion nodes.
  3. Normalization & Enrichment: Data is processed, normalized, and enriched.
  4. Indexing: Processed data is indexed into the Elasticsearch datastore.
  5. Analysis: ML models and detection rules analyze indexed data in real-time.
  6. Visualization: Results are visualized in dashboards and alerts are triggered if threats are detected.

Scaling Strategy

Scaling DimensionApproach
IngestionAdd more ingestion nodes horizontally.
ProcessingScale out processors to handle high EPS (events per second).
StorageExpand Elasticsearch clusters with more nodes.
AnalyticsDeploy additional ML model servers for parallel analysis.
VisualizationUse load balancers for scaling CyberDoyen Web UI instances.

Why CyberDoyen SIEM?

  • Performance at Scale: Supports up to 50,000 EPS (events per second) on a standard 3-node cluster.
  • Flexible Architecture: Easily adapt to new log sources, detection techniques, and scaling requirements.
  • Rapid Threat Detection: Advanced machine learning and rule-based detection ensure faster response times.
  • Customizable Dashboards: Visualize only the information that matters most to your team.

Ready to dive deeper?
See the Installation Guide and Performance Tuning Guide for next steps.

Last updated on