Skip to main content
Understand the Honeycomb Private Cloud technical architecture.
This page provides a high-level overview of Private Cloud architecture. Detailed deployment and networking guidance will be provided closer to deployment.
Honeycomb Private Cloud runs on the same architecture that powers Honeycomb’s SaaS platform, delivering high-speed query performance, AI-native intelligence, and scalable telemetry processing within your AWS environment.

​
Architecture Principles

Honeycomb Private Cloud uses a distributed, service-oriented architecture designed for performance, scalability, and reliability:
  • Horizontally scalable: Components scale by adding instances rather than increasing individual resource size
  • High availability: Resources distributed across multiple AWS Availability Zones (AZs) for resilience
  • Service separation: Distinct components for ingestion, query processing, storage, and user interface
  • Real-time processing: Event data available for query immediately upon ingestion

​
Core Components

Your Private Cloud deployment includes:
  • Ingestion pipeline: Receives and processes telemetry data from your applications
  • Query engine: Honeycomb’s custom column store for high-speed, high-cardinality queries
  • Storage layer: Durable persistence of telemetry events
  • Web application: Honeycomb UI for exploration, analysis, and visualization
  • AI services: Canvas and MCP for intelligent telemetry exploration
  • API services: Programmatic access for integrations and automation
  • Metadata services: User management, authentication, and configuration

​
AWS Infrastructure

Honeycomb Private Cloud leverages multiple AWS services:
  • Compute: EC2, EKS, ECS, Lambda
  • Storage: S3, RDS
  • Caching: ElastiCache (Redis, Memcached)
  • Networking: VPC, subnets, load balancers
  • Configuration: SSM Parameter Store
  • High Availability: Multi-AZ, automatic failover, redundant caching
  • AI: Bedrock
This list is not exhaustive.

​
Infrastructure Requirements

To ensure Honeycomb Private Cloud runs smoothly, you need infrastructure that matches your telemetry volume, query patterns, and availability needs.

​
Resource Sizing

Sizing depends on your telemetry volume and query patterns. Honeycomb provides sizing recommendations based on:
  • Expected events per second: Peak and average telemetry ingestion rate
  • Concurrent users: Number of people actively using Honeycomb
  • Query patterns: Frequency and complexity of queries
  • Data retention: How long telemetry data is kept
  • Growth projections: Expected increase in telemetry volume
Work with your Honeycomb account team during planning to determine appropriate infrastructure sizing for your deployment.

​
Inbound Traffic

Your Honeycomb deployment receives:
  • Telemetry data from your applications and services
  • User traffic to the Honeycomb UI (web browser access)
  • API requests from integrations and automation

​
Outbound Traffic

Your Honeycomb deployment connects to:
  • Your authentication provider (SAML/Okta IdP)
  • Your email gateway (for notifications)
  • Your Slack workspace (if using Slack integration)

​
Integration Requirements

Honeycomb Private Cloud connects with your existing systems for secure access, alerts, and team collaboration. Key integrations include:
  • Authentication: SAML 2.0, ADFS, other SAML 2.0-compliant IdPs
  • Email notifications: SMTP relay or email gateway
  • Slack (optional): Custom app installation and OAuth token
  • GitHub (optional): Custom GitHub app for deployment gates

​
Deployment-Specific Requirements

Detailed deployment and networking guidance will be provided closer to deployment. To learn more about operational responsibilities, visit Deployment Models.

​
Limitations

To guide your deployment decisions, note these constraints:
  • AWS GovCloud regions: AI model access differs, which may affect how Canvas and MCP features work.