Collecting and Exporting EF Core Performance Metrics with OpenTelemetry

Monitoring application performance is essential for catching database bottlenecks before they impact users. Entity Framework Core (EF Core) exposes performance data natively through the standard .NET System.Diagnostics.Metrics API under the meter name Microsoft.EntityFrameworkCore.

By integrating this meter with OpenTelemetry, you can automatically track active database connections, query volumes, and compiled query cache performance across your application.

Step-by-Step Implementation

1.Install Required Packages:Terminal command.

Run the following command to add the OpenTelemetry hosting and console exporter packages to your project:

Bash

dotnet add package OpenTelemetry.Extensions.Hosting
dotnet add package OpenTelemetry.Exporter.Console

Verification: Check your .csproj file to confirm that OpenTelemetry.Extensions.Hosting and the exporter packages are successfully listed under <ItemGroup>.

2.Configure OpenTelemetry Metrics in Program.cs:Code setup.

Register OpenTelemetry metrics during application startup and subscribe to the EF Core meter using .AddMeter("Microsoft.EntityFrameworkCore"):

C#

using OpenTelemetry.Metrics;

var builder = WebApplication.CreateBuilder(args);

// Configure OpenTelemetry Metrics
builder.Services.AddOpenTelemetry()
    .WithMetrics(metrics => metrics
        .AddAspNetCoreInstrumentation() // Captures HTTP performance metrics
        .AddRuntimeInstrumentation()     // Captures .NET runtime metrics (memory, CPU)
        .AddMeter("Microsoft.EntityFrameworkCore") // Captures EF Core built-in metrics
        .AddConsoleExporter());          // Exports metrics to the console for local verification

var app = builder.Build();
// ... configure middleware and endpoints
app.Run();

Verification: Run dotnet build to verify that your project compiles cleanly without missing namespace references.

3.Verify Metric Collection:Runtime check.

Start your application and trigger a few API endpoints that execute database operations.

Verification: Inspect your application console output to confirm that performance counters like query counts and active contexts are actively streaming.

Key Metrics Tracked by EF Core

Once configured, EF Core automatically feeds the following performance measurements into your OpenTelemetry pipeline:

Metric Name	Instrument Type	Description
microsoft.entityframeworkcore.active_dbcontexts	ObservableUpDownCounter	Tracks the number of currently active DbContext instances.
microsoft.entityframeworkcore.queries	ObservableCounter	Cumulative count of total database queries executed.
microsoft.entityframeworkcore.savechanges	ObservableCounter	Cumulative count of database save operations executed.
microsoft.entityframeworkcore.compiled_query_cache_hits	ObservableCounter	Tracks cache hits for compiled queries.
microsoft.entityframeworkcore.compiled_query_cache_misses	ObservableCounter	Tracks cache misses for compiled queries.

Production Setup: Exporting to Prometheus

In production environments, you typically replace the console exporter with a Prometheus exporter to visualize your metrics in Grafana:

Bash

dotnet add package OpenTelemetry.Exporter.Prometheus.AspNetCore

Update your metrics registration in Program.cs:

C#

builder.Services.AddOpenTelemetry()
    .WithMetrics(metrics => metrics
        .AddMeter("Microsoft.EntityFrameworkCore")
        .AddPrometheusExporter()); // Exposes a /metrics scraping endpoint

// Map the scraping endpoint for Prometheus
app.MapPrometheusScrapingEndpoint();