Skip to main content

LogCentral: Simple Architecture Overview

Updated over 2 weeks ago

How it Works — The Basics

  1. Log Sources to Ingestion
    Your systems (servers, routers, wireless access points) emit logs. These syslogs are sent to LogCentral.

  2. Ingestion & Buffering
    Incoming logs are buffered or queued so that sudden bursts don’t overwhelm the system. Basic validation and normalization happen here.

  3. Tenant Routing & Processing
    Each log is assigned to a specific customer (organization / location) context.

  4. Storage Layers
    Logs are stored in different tiers depending on their age and how often they’re accessed:

    • Hot (recent logs): stored for fast querying

    • Cold / Archive: compressed, less frequently accessed, long-term retention

  5. Query, Visualization
    Users access logs via UI & can search / download them.

  6. Retention & Deletion
    When logs exceed retention policies, they’re purged & deleted per configured rules.

Key Principles & Qualities

  • Multi-tenant, isolated
    Though the infrastructure is shared, each customer’s data and operations are isolated logically: one cannot access another’s logs or settings.

  • Scalable & resilient
    The system is built to scale horizontally: new nodes can be added to handle growth. Redundancy ensures high availability (e.g. replicated storage, failover).

  • Secure & controlled
    Encryption in transit and at rest, role-based permissions, and monitoring are standard. Only authorized users and services access data.

  • Cost-efficient tiering
    By moving logs across storage tiers (hot → cold), we balance performance vs cost: recent logs stay fast, older ones move to cheaper storage.

Did this answer your question?