The Case for Self-Hosted Observability

In the modern DevOps landscape, observability is not merely an optional feature; it is the backbone of operational excellence. For years, Datadog has been the industry standard, offering a comprehensive suite of tools. However, as organizations scale, the pricing model—often tied to data ingestion volumes—can lead to unpredictable and astronomical monthly expenses. This has triggered a growing trend: the migration toward open-source, self-hosted alternatives like SigNoz.

SigNoz provides a full-stack observability solution, combining metrics, logs, and traces into a single, cohesive pane of glass. By deploying this on your own Virtual Private Server (VPS), you reclaim control over your data while drastically reducing vendor lock-in and infrastructure costs.

Why Transition from Datadog to SigNoz?

The primary driver for many enterprises is cost optimization. While Datadog is convenient, its cost scales linearly with your infrastructure growth. Conversely, SigNoz allows you to scale based on your hardware capacity, meaning you pay for the VPS resources rather than a tax on your telemetry data.

Data Sovereignty: Keep sensitive operational data within your private infrastructure.
Unified Interface: SigNoz is built on OpenTelemetry, ensuring seamless integration across various languages and frameworks.
Customizability: Since it is open-source, your engineering team can extend features to meet specific business needs that proprietary platforms might restrict.
Simplified Troubleshooting: Correlating traces with logs and metrics is natively integrated, reducing the time spent jumping between different tool tabs.

Architecture Overview

Deploying SigNoz on a VPS requires a robust understanding of the underlying stack. SigNoz utilizes several core components:

Collector (OpenTelemetry): The industry-standard agent that collects traces, metrics, and logs from your applications.
Query Service: The engine that processes and serves data to the frontend.
ClickHouse: The backbone of SigNoz, a high-performance, column-oriented database designed for real-time analytics.
Frontend: A React-based UI that provides intuitive visualization and alerting capabilities.

Pro Tip: When choosing a VPS for your observability stack, prioritize high I/O performance and sufficient RAM. Because ClickHouse performs heavy write operations, NVMe-backed storage is strongly recommended for production environments.

Step-by-Step Migration Strategy

1. Assessing Current Telemetry

Before moving, audit your current Datadog instrumentation. Determine which metrics and traces are mission-critical. Since SigNoz is OpenTelemetry-native, most of your existing instrumentation can be migrated by simply changing the OTLP exporter endpoints.

2. Infrastructure Provisioning

Provision your VPS instances. We recommend a distributed setup for high availability, even for self-hosted solutions. Ensure your security groups only allow traffic from your application clusters to the SigNoz collector ports.

3. Deployment with Docker/Kubernetes

SigNoz provides official installation scripts for both Docker and Kubernetes. For most VPS setups, the Docker Compose approach is the fastest route to a functional POC (Proof of Concept).

git clone [https://github.com/SigNoz/signoz.git](https://github.com/SigNoz/signoz.git)
cd signoz/deploy/
./install.sh

4. Redirecting Traffic

Update your application sidecars or agents to point to your new SigNoz collector endpoint. Begin by routing non-critical logs to verify data integrity before moving your core performance traces.

Overcoming Challenges

While the benefits are significant, self-hosting requires operational rigor. Unlike a SaaS provider, you are responsible for the uptime of your observability stack. We recommend implementing standard monitoring practices for the monitor itself (meta-monitoring) to ensure your observability platform does not experience downtime during a critical production incident.

Additionally, consider your retention strategy. ClickHouse is highly efficient, but unlimited data storage will eventually exhaust your VPS disks. Configure appropriate TTL (Time-To-Live) settings within your database schema to balance historical data needs with hardware limitations.

Conclusion

Transitioning from Datadog to SigNoz is more than just a cost-saving measure; it is a strategic investment in infrastructure control. By leveraging the power of OpenTelemetry and the performance of ClickHouse, organizations can build a high-fidelity observability platform that scales with their requirements rather than their budget. While it requires an upfront investment in engineering effort, the long-term gains in data autonomy and financial predictability are well worth the transition.

Migrating from Datadog to SigNoz: Building a Robust Full-Stack Observability Platform on VPS