Observability to Azure Monitor Migration Center¶

The definitive resource for migrating from Datadog, New Relic, Splunk Observability, and Dynatrace to Azure Monitor, Application Insights, and Azure Managed Grafana.

Who this is for¶

This migration center serves platform engineers, SREs, DevOps leads, IT directors, application developers, and federal technology leaders who are evaluating or executing a migration from third-party observability platforms to Azure Monitor. Whether you are responding to cost pressure from per-host licensing, consolidating observability onto the Azure control plane, addressing federal compliance requirements that third-party vendors cannot meet, or unifying application and infrastructure monitoring, these resources provide the evidence, patterns, and step-by-step guidance to execute confidently.

Quick-start decision matrix¶

Decision: Azure Monitor vs Azure Managed Grafana vs Hybrid¶

Before migrating, decide on your target architecture. Azure Monitor supports three deployment models.

Model 1: Azure Monitor Native¶

Best for: Teams standardizing on Azure, new to observability tooling, or prioritizing simplicity and cost.

• Dashboards: Azure Workbooks
• Alerting: Azure Monitor Alerts
• Query language: KQL
• Cost: Lowest (no additional Grafana compute)
• Strengths: Deepest Azure integration, simplest operations, single billing

Model 2: Azure Managed Grafana + Azure Monitor Backend¶

Best for: Teams with existing Grafana expertise, multi-cloud environments, or strong PromQL/Grafana dashboard investment.

• Dashboards: Azure Managed Grafana (fully managed)
• Data sources: Azure Monitor, Azure Data Explorer, Prometheus, Log Analytics
• Query language: KQL + PromQL
• Cost: Moderate (adds Grafana instance cost)
• Strengths: Familiar Grafana UX, community dashboard ecosystem, multi-source correlation

Model 3: Hybrid (Workbooks + Grafana)¶

Best for: Large organizations with mixed teams -- platform teams use Workbooks for Azure infrastructure, application teams use Grafana for microservices.

• Azure infrastructure: Workbooks (VM Insights, Container Insights, network)
• Application observability: Grafana (APM dashboards, Prometheus metrics, custom panels)
• Cost: Highest (both tools maintained)
• Strengths: Maximum flexibility, best-of-both-worlds for diverse teams

flowchart TD
    A["Your Team's Background"] --> B{"Existing Grafana<br/>expertise?"}
    B -->|Yes| C{"Multi-cloud or<br/>Azure-primary?"}
    B -->|No| D["Model 1:<br/>Azure Monitor Native"]
    C -->|Multi-cloud| E["Model 2:<br/>Managed Grafana + Azure Monitor"]
    C -->|Azure-primary| F{"Need both infra<br/>+ app teams?"}
    F -->|Yes| G["Model 3: Hybrid"]
    F -->|No| E

Strategic resources¶

These documents provide the business case, cost analysis, and strategic framing for decision-makers.

Migration guides¶

Domain-specific deep dives covering every aspect of an observability migration.

Tutorials¶

Hands-on walkthroughs for the most common migration tasks.

Technical references¶

Government and federal¶

How CSA-in-a-Box fits¶

Azure Monitor is CSA-in-a-Box's native observability layer. Every component deployed by the reference architecture emits diagnostics to Azure Monitor.

flowchart LR
    subgraph CSA["CSA-in-a-Box Components"]
        Fabric["Microsoft Fabric"]
        DBX["Databricks"]
        ADF["Azure Data Factory"]
        Purview["Microsoft Purview"]
        AOAI["Azure OpenAI"]
        PBI["Power BI"]
        KV["Key Vault"]
        ADLS["ADLS Gen2 / OneLake"]
    end

    subgraph Monitor["Azure Monitor"]
        LA["Log Analytics Workspace"]
        AI["Application Insights"]
        Metrics["Azure Monitor Metrics"]
        Alerts["Alert Rules"]
        WB["Workbooks"]
    end

    Fabric --> LA
    DBX --> LA
    ADF --> LA
    Purview --> LA
    AOAI --> LA
    AOAI --> AI
    PBI --> LA
    KV --> LA
    ADLS --> LA

    LA --> Alerts
    AI --> Alerts
    Metrics --> Alerts
    LA --> WB
    AI --> WB
    Metrics --> WB

Migrating to Azure Monitor does not just replace a third-party observability tool -- it unifies application monitoring with the platform monitoring that CSA-in-a-Box already provides. Benefits include:

• Single Log Analytics workspace for both platform diagnostics (Fabric capacity, ADF pipeline runs, Purview scan health) and application telemetry (traces, custom metrics, exceptions)
• Cross-component correlation using KQL to join application errors with infrastructure events (e.g., correlate a spike in API errors with a Databricks cluster autoscale event)
• Unified alerting with action groups that notify the same on-call teams for both platform and application issues
• Power BI integration for business observability dashboards that combine application performance metrics with data pipeline SLAs
• Cost visibility through Azure Cost Management -- observability spend is visible alongside compute, storage, and data platform costs

Migration timeline overview¶

gantt
    title Typical Observability Migration (20 weeks)
    dateFormat  YYYY-MM-DD
    section Discovery
    Inventory agents, dashboards, alerts       :a1, 2026-05-01, 2w
    section Foundation
    Deploy Log Analytics + App Insights        :a2, after a1, 3w
    section Dual-Ship
    Logs + metrics to both platforms            :a3, after a2, 5w
    section APM
    Application Insights instrumentation       :a4, after a2, 6w
    section Validation
    Parallel run + alert parity                :a5, after a3, 4w
    section Cutover
    Decommission vendor agents + licenses      :a6, after a5, 2w
    section Optimization
    Tune sampling, tiers, cost                 :a7, after a6, 4w

Recommended migration sequence¶

For organizations migrating from any of the three major observability platforms, follow this recommended reading order:

1. Business case: Why Azure Monitor then TCO Analysis
2. Feature parity: Complete Feature Mapping
3. Plan by domain:
   • APM Migration -- for application performance monitoring
   • Log Migration -- for centralized logging
   • Metrics Migration -- for custom and infrastructure metrics
   • Alerting Migration -- for alert rules and on-call
   • Dashboard Migration -- for visualization
4. Hands-on:
   • Tutorial: Application Insights
   • Tutorial: Log Analytics
5. Federal (if applicable): Federal Migration Guide
6. Validate: Benchmarks
7. Operationalize: Best Practices

Last updated: 2026-04-30 Maintainers: CSA-in-a-Box core team Related: Migration Playbook | Splunk to Sentinel | CSA-in-a-Box Observability Patterns | Monitoring Best Practices