Merge pull request #2 from JerrettDavis/feature/add-enterprise-features

feat: add enterprise resilience features and comprehensive documentation

Author: JerrettDavis

ExperimentFramework.slnx

@@ -7,6 +7,8 @@
   <Folder Name="/src/">
     <Project Path="benchmarks/ExperimentFramework.Benchmarks/ExperimentFramework.Benchmarks.csproj" />
     <Project Path="src/ExperimentFramework.Generators/ExperimentFramework.Generators.csproj" />
+    <Project Path="src/ExperimentFramework.Metrics.Exporters/ExperimentFramework.Metrics.Exporters.csproj" />
+    <Project Path="src/ExperimentFramework.Resilience/ExperimentFramework.Resilience.csproj" />
     <Project Path="src/ExperimentFramework/ExperimentFramework.csproj" />
   </Folder>
   <Folder Name="/tests/">

README.md

@@ -1,31 +1,30 @@
 # ExperimentFramework
 
-A .NET framework for runtime-switchable A/B testing, feature flags, trial fallback, and comprehensive observability.
+A .NET library for routing service calls through configurable trials based on feature flags, configuration values, or custom routing logic.
 
-**Version 0.1.0** - Production-ready with source-generated zero-overhead proxies
+## Features
 
-## Key Features
-
-**Multiple Selection Modes**
+**Selection Modes**
 - Boolean feature flags (`true`/`false` keys)
 - Configuration values (string variants)
 - Variant feature flags (IVariantFeatureManager integration)
-- Sticky routing (deterministic user/session-based A/B testing)
+- Sticky routing (deterministic user/session-based routing)
+
+**Resilience**
+- Timeout enforcement with fallback
+- Circuit breaker (Polly integration)
+- Kill switch for disabling experiments at runtime
 
-**Enterprise Observability**
-- OpenTelemetry distributed tracing support
+**Observability**
+- OpenTelemetry tracing
+- Metrics collection (Prometheus, OpenTelemetry)
 - Built-in benchmarking and error logging
-- Zero overhead when telemetry disabled
 
-**Flexible Configuration**
-- Custom naming conventions
+**Configuration**
 - Error policies with fallback strategies
+- Custom naming conventions
 - Decorator pipeline for cross-cutting concerns
-
-**Type-Safe & DI-Friendly**
-- Composition-root driven registration
-- Full dependency injection integration
-- Strongly-typed builder API
+- Dependency injection integration
 
 ## Quick Start
 
@@ -230,6 +229,98 @@ Tries ordered list of fallback trials:
 // Fine-grained control over fallback strategy
 ```
 
+## Timeout Enforcement
+
+Prevent slow trials from degrading system performance:
+
+```csharp
+var experiments = ExperimentFrameworkBuilder.Create()
+    .Define<IMyDatabase>(c => c
+        .UsingFeatureFlag("UseCloudDb")
+        .AddDefaultTrial<LocalDb>("false")
+        .AddTrial<CloudDb>("true")
+        .OnErrorRedirectAndReplayDefault())
+    .WithTimeout(TimeSpan.FromSeconds(5), TimeoutAction.FallbackToDefault)
+    .UseDispatchProxy();
+```
+
+**Actions:**
+- `TimeoutAction.ThrowException` - Throw `TimeoutException` when trial exceeds timeout
+- `TimeoutAction.FallbackToDefault` - Automatically fallback to default trial on timeout
+
+See [Timeout Enforcement Guide](docs/user-guide/timeout-enforcement.md) for detailed examples.
+
+## Circuit Breaker
+
+Automatically disable failing trials using Polly:
+
+```bash
+dotnet add package ExperimentFramework.Resilience
+```
+
+```csharp
+var experiments = ExperimentFrameworkBuilder.Create()
+    .Define<IMyService>(c => c
+        .UsingFeatureFlag("UseNewService")
+        .AddDefaultTrial<StableService>("false")
+        .AddTrial<NewService>("true")
+        .OnErrorRedirectAndReplayDefault())
+    .WithCircuitBreaker(options =>
+    {
+        options.FailureRatioThreshold = 0.5;      // Open after 50% failure rate
+        options.MinimumThroughput = 10;            // Need 10 calls to assess
+        options.SamplingDuration = TimeSpan.FromSeconds(30);
+        options.BreakDuration = TimeSpan.FromSeconds(60);
+        options.OnCircuitOpen = CircuitBreakerAction.FallbackToDefault;
+    })
+    .UseDispatchProxy();
+```
+
+See [Circuit Breaker Guide](docs/user-guide/circuit-breaker.md) for advanced configuration.
+
+## Metrics Collection
+
+Track experiment performance with Prometheus or OpenTelemetry:
+
+```bash
+dotnet add package ExperimentFramework.Metrics.Exporters
+```
+
+```csharp
+var prometheusMetrics = new PrometheusExperimentMetrics();
+var experiments = ExperimentFrameworkBuilder.Create()
+    .Define<IMyService>(c => c.UsingFeatureFlag("MyFeature")...)
+    .WithMetrics(prometheusMetrics)
+    .UseDispatchProxy();
+
+app.MapGet("/metrics", () => prometheusMetrics.GeneratePrometheusOutput());
+```
+
+**Collected Metrics:**
+- `experiment_invocations_total` (counter) - Total invocations per experiment/trial
+- `experiment_duration_seconds` (histogram) - Duration of each invocation
+
+See [Metrics Guide](docs/user-guide/metrics.md) for OpenTelemetry integration and Grafana dashboards.
+
+## Kill Switch
+
+Emergency shutdown for problematic experiments:
+
+```csharp
+var killSwitch = new InMemoryKillSwitchProvider();
+
+var experiments = ExperimentFrameworkBuilder.Create()
+    .Define<IMyDatabase>(c => c.UsingFeatureFlag("UseCloudDb")...)
+    .WithKillSwitch(killSwitch)
+    .UseDispatchProxy();
+
+// Emergency disable
+killSwitch.DisableExperiment(typeof(IMyDatabase));
+killSwitch.DisableTrial(typeof(IMyDatabase), "cloud");
+```
+
+See [Kill Switch Guide](docs/user-guide/kill-switch.md) for distributed scenarios with Redis.
+
 ## Custom Naming Conventions
 
 Replace default selector naming:

benchmarks/ExperimentFramework.Benchmarks/ProxyOverheadBenchmarks.cs

@@ -247,7 +247,7 @@ public string[] Direct_RepeatedSync_10Calls()
         using var scope = _directServiceProvider.CreateScope();
         var service = scope.ServiceProvider.GetRequiredService<ISimpleService>();
         var results = new string[10];
-        for (int i = 0; i < 10; i++)
+        for (var i = 0; i < 10; i++)
         {
             results[i] = service.GetValue();
         }
@@ -260,7 +260,7 @@ public string[] Proxied_Configuration_RepeatedSync_10Calls()
         using var scope = _proxiedConfigurationServiceProvider.CreateScope();
         var service = scope.ServiceProvider.GetRequiredService<ISimpleService>();
         var results = new string[10];
-        for (int i = 0; i < 10; i++)
+        for (var i = 0; i < 10; i++)
         {
             results[i] = service.GetValue();
         }
@@ -273,7 +273,7 @@ public string[] Direct_RepeatedSync_100Calls()
         using var scope = _directServiceProvider.CreateScope();
         var service = scope.ServiceProvider.GetRequiredService<ISimpleService>();
         var results = new string[100];
-        for (int i = 0; i < 100; i++)
+        for (var i = 0; i < 100; i++)
         {
             results[i] = service.GetValue();
         }
@@ -286,7 +286,7 @@ public string[] Proxied_Configuration_RepeatedSync_100Calls()
         using var scope = _proxiedConfigurationServiceProvider.CreateScope();
         var service = scope.ServiceProvider.GetRequiredService<ISimpleService>();
         var results = new string[100];
-        for (int i = 0; i < 100; i++)
+        for (var i = 0; i < 100; i++)
         {
             results[i] = service.GetValue();
         }
@@ -299,7 +299,7 @@ public async Task<string[]> Direct_RepeatedAsync_10Calls()
         using var scope = _directServiceProvider.CreateScope();
         var service = scope.ServiceProvider.GetRequiredService<ISimpleService>();
         var results = new string[10];
-        for (int i = 0; i < 10; i++)
+        for (var i = 0; i < 10; i++)
         {
             results[i] = await service.GetValueAsync();
         }
@@ -312,7 +312,7 @@ public async Task<string[]> Proxied_Configuration_RepeatedAsync_10Calls()
         using var scope = _proxiedConfigurationServiceProvider.CreateScope();
         var service = scope.ServiceProvider.GetRequiredService<ISimpleService>();
         var results = new string[10];
-        for (int i = 0; i < 10; i++)
+        for (var i = 0; i < 10; i++)
         {
             results[i] = await service.GetValueAsync();
         }
@@ -325,7 +325,7 @@ public async Task<string[]> Direct_RepeatedAsync_100Calls()
         using var scope = _directServiceProvider.CreateScope();
         var service = scope.ServiceProvider.GetRequiredService<ISimpleService>();
         var results = new string[100];
-        for (int i = 0; i < 100; i++)
+        for (var i = 0; i < 100; i++)
         {
             results[i] = await service.GetValueAsync();
         }
@@ -338,7 +338,7 @@ public async Task<string[]> Proxied_Configuration_RepeatedAsync_100Calls()
         using var scope = _proxiedConfigurationServiceProvider.CreateScope();
         var service = scope.ServiceProvider.GetRequiredService<ISimpleService>();
         var results = new string[100];
-        for (int i = 0; i < 100; i++)
+        for (var i = 0; i < 100; i++)
         {
             results[i] = await service.GetValueAsync();
         }

benchmarks/ExperimentFramework.Benchmarks/RealWorldScenarioBenchmarks.cs

@@ -259,7 +259,7 @@ public string[] Direct_CPUBound_Repeated_10Calls()
         using var scope = _directCacheProvider.CreateScope();
         var cache = scope.ServiceProvider.GetRequiredService<ICache>();
         var results = new string[10];
-        for (int i = 0; i < 10; i++)
+        for (var i = 0; i < 10; i++)
         {
             results[i] = cache.ComputeHash($"test-data-{i}");
         }
@@ -272,7 +272,7 @@ public string[] Proxied_CPUBound_Repeated_10Calls()
         using var scope = _proxiedCacheProvider.CreateScope();
         var cache = scope.ServiceProvider.GetRequiredService<ICache>();
         var results = new string[10];
-        for (int i = 0; i < 10; i++)
+        for (var i = 0; i < 10; i++)
         {
             results[i] = cache.ComputeHash($"test-data-{i}");
         }
@@ -285,7 +285,7 @@ public string[] Direct_CPUBound_Repeated_100Calls()
         using var scope = _directCacheProvider.CreateScope();
         var cache = scope.ServiceProvider.GetRequiredService<ICache>();
         var results = new string[100];
-        for (int i = 0; i < 100; i++)
+        for (var i = 0; i < 100; i++)
         {
             results[i] = cache.ComputeHash($"test-data-{i}");
         }
@@ -298,7 +298,7 @@ public string[] Proxied_CPUBound_Repeated_100Calls()
         using var scope = _proxiedCacheProvider.CreateScope();
         var cache = scope.ServiceProvider.GetRequiredService<ICache>();
         var results = new string[100];
-        for (int i = 0; i < 100; i++)
+        for (var i = 0; i < 100; i++)
         {
             results[i] = cache.ComputeHash($"test-data-{i}");
         }
@@ -311,7 +311,7 @@ public string[] Proxied_CPUBound_Repeated_100Calls()
         using var scope = _directDatabaseProvider.CreateScope();
         var db = scope.ServiceProvider.GetRequiredService<IDatabase>();
         var results = new Customer?[10];
-        for (int i = 0; i < 10; i++)
+        for (var i = 0; i < 10; i++)
         {
             results[i] = await db.GetCustomerByIdAsync(1);
         }
@@ -324,7 +324,7 @@ public string[] Proxied_CPUBound_Repeated_100Calls()
         using var scope = _proxiedDatabaseProvider.CreateScope();
         var db = scope.ServiceProvider.GetRequiredService<IDatabase>();
         var results = new Customer?[10];
-        for (int i = 0; i < 10; i++)
+        for (var i = 0; i < 10; i++)
         {
             results[i] = await db.GetCustomerByIdAsync(1);
         }
@@ -337,7 +337,7 @@ public string[] Proxied_CPUBound_Repeated_100Calls()
         using var scope = _directDatabaseProvider.CreateScope();
         var db = scope.ServiceProvider.GetRequiredService<IDatabase>();
         var results = new Customer?[100];
-        for (int i = 0; i < 100; i++)
+        for (var i = 0; i < 100; i++)
         {
             results[i] = await db.GetCustomerByIdAsync(1);
         }
@@ -350,7 +350,7 @@ public string[] Proxied_CPUBound_Repeated_100Calls()
         using var scope = _proxiedDatabaseProvider.CreateScope();
         var db = scope.ServiceProvider.GetRequiredService<IDatabase>();
         var results = new Customer?[100];
-        for (int i = 0; i < 100; i++)
+        for (var i = 0; i < 100; i++)
         {
             results[i] = await db.GetCustomerByIdAsync(1);
         }