Faced with the reality that no single tool covers the full spectrum of monitoring + observability, CIOs have two options : hunt for the unicorn (a single tool that does everything well) or embrace complementarity.
Our conviction is firm : the first option is a fantasy. The second, when well orchestrated, delivers excellent results. This is what we call the two-pole approach.
The Guiding Principle
| UNIFIED OBSERVABILITY STRATEGY | |
|---|---|
| INFRASTRUCTURE POLE “From the network layer to the system” • Network, Storage, Hypervisors • Physical servers and VMs • Cloud (Azure, AWS) SNMP, WMI, API · PRTG, Zabbix, Centreon | OBSERVABILITY POLE “From the system to the applications” • Applications and microservices • Containers and K8s • User experience OpenTelemetry (OTLP) · VictoriaMetrics, Grafana |
| INTEGRATION : Grafana (unified dashboards) · Centralized alerting · Cross-layer correlation | |
Why This Approach?
- Each tool excels in its own domain : you capitalize on the strengths of each rather than accepting the trade-offs of a solution that is merely average at everything.
- Gradual transition : you keep the existing setup that works and add the observability building block over time. No big bang, no disruption.
- Open standards : OpenTelemetry as a common language, no vendor lock-in. If you swap out a tool tomorrow, the instrumentation stays in place.
- Unified visibility : Grafana brings together data from both poles in a single interface.
- Cost control : total control over the volumes ingested and stored, with no surprises tied to volume-based pricing.
The Overlap Zone, and the Convergence of Logs
The two poles are not airtight. There is a legitimate overlap zone around the system layer. This redundancy—often seen as a flaw—is in fact an asset : it guarantees continuity of visibility, even when one of the poles is under maintenance.
The most significant convergence plays out around logs. Historically, managing infrastructure logs was complex : monitoring tools didn’t ingest them, and log management solutions (syslog, ELK) operated in silos. Today, both poles can process logs, creating a natural bridge between monitoring and observability.
Centreon, for example, launched its Centreon Monitoring Agent (CMA) in 2025, built on the OpenTelemetry protocol—a strong signal of this convergence : the monitoring world is adopting the standards of observability.
What About the Teams?
The two-pole approach doesn’t just separate technical tools. It reflects an organizational reality : network and system administrators work in the world of monitoring, while DevOps and SREs work in the world of observability.
Forcing these two groups into the same tool satisfies no one. Giving each of them the tool suited to their role, federated by a common layer (Grafana), respects their expertise while building the big-picture view.
In the next article, we’ll detail the open-source stack that makes this approach concrete : OpenTelemetry, VictoriaMetrics, Grafana.
This article is drawn from our white paper “From Monitoring to Observability” (PDF, 2026).
