Add .NET select-algorithm sample

ai/select-algorithm-dotnet/.gitignore

@@ -0,0 +1,11 @@
+# Build output
+bin/
+obj/
+
+# User settings
+*.user
+*.suo
+
+# Environment
+.env
+appsettings.Development.json

ai/select-algorithm-dotnet/Program.cs

@@ -0,0 +1,98 @@
+using Microsoft.Extensions.Configuration;
+using Microsoft.Extensions.DependencyInjection;
+using Microsoft.Extensions.Logging;
+using SelectAlgorithm.Services;
+using SelectAlgorithm.Utilities;
+using System.Reflection;
+
+namespace SelectAlgorithm;
+
+class Program
+{
+    static async Task Main(string[] args)
+    {
+        var configuration = new ConfigurationBuilder()
+            .SetBasePath(Directory.GetCurrentDirectory())
+            .AddJsonFile("appsettings.json", optional: false, reloadOnChange: true)
+            .AddEnvironmentVariables()
+            .Build();
+
+        var services = new ServiceCollection()
+            .AddLogging(builder => builder
+                .AddConsole()
+                .SetMinimumLevel(LogLevel.Information))
+            .AddSingleton<IConfiguration>(configuration);
+
+        var serviceProvider = services.BuildServiceProvider();
+        var logger = serviceProvider.GetRequiredService<ILogger<Program>>();
+
+        try
+        {
+            var openAiEndpoint = configuration["AZURE_OPENAI_EMBEDDING_ENDPOINT"]
+                ?? throw new InvalidOperationException("AZURE_OPENAI_EMBEDDING_ENDPOINT not configured. Set as environment variable or in appsettings.json.");
+            var openAiModel = configuration["AZURE_OPENAI_EMBEDDING_MODEL"]
+                ?? throw new InvalidOperationException("AZURE_OPENAI_EMBEDDING_MODEL not configured. Set as environment variable or in appsettings.json.");
+            var mongoClusterName = configuration["MONGO_CLUSTER_NAME"]
+                ?? throw new InvalidOperationException("MONGO_CLUSTER_NAME not configured. Set as environment variable or in appsettings.json.");
+            var tenantId = configuration["AZURE_TENANT_ID"]
+                ?? throw new InvalidOperationException("AZURE_TENANT_ID not configured. Set as environment variable or in appsettings.json.");
+
+            var databaseName = configuration["DatabaseName"] ?? "Hotels";
+            var embeddedField = configuration["EmbeddedField"] ?? "DescriptionVector";
+            if (!int.TryParse(configuration["EmbeddingDimensions"] ?? "1536", out int embeddingDimensions))
+                throw new InvalidOperationException("EmbeddingDimensions must be a valid integer");
+            if (!int.TryParse(configuration["LoadBatchSize"] ?? "100", out int loadBatchSize))
+                throw new InvalidOperationException("LoadBatchSize must be a valid integer");
+            var searchQuery = configuration["SearchQuery"] ?? "quintessential lodging near running trails, eateries, retail";
+            if (!int.TryParse(configuration["TopK"] ?? "5", out int topK))
+                throw new InvalidOperationException("TopK must be a valid integer");
+
+            var dataFileRelative = Environment.GetEnvironmentVariable("DATA_FILE_WITH_VECTORS") ?? "../../data/Hotels_Vector.json";
+            var assemblyDir = Path.GetDirectoryName(Assembly.GetExecutingAssembly().Location) ?? string.Empty;
+            var dataFilePath = Path.GetFullPath(Path.Combine(assemblyDir, dataFileRelative));
+
+            var algorithmFilter = (Environment.GetEnvironmentVariable("ALGORITHM") ?? "all").Trim().ToLower();
+            var similarityFilter = (Environment.GetEnvironmentVariable("SIMILARITY") ?? "COS").Trim().ToUpper();
+
+            logger.LogInformation("Initializing clients with passwordless authentication...");
+            var (aiClient, dbClient) = Utils.GetClientsPasswordless(openAiEndpoint, mongoClusterName, tenantId);
+
+            var comparisonLogger = serviceProvider.GetRequiredService<ILoggerFactory>()
+                .CreateLogger<VectorComparisonService>();
+
+            var comparisonService = new VectorComparisonService(
+                comparisonLogger,
+                aiClient,
+                dbClient,
+                databaseName,
+                embeddedField,
+                embeddingDimensions,
+                loadBatchSize,
+                topK
+            );
+
+            var results = await comparisonService.RunComparisonAsync(
+                dataFilePath,
+                searchQuery,
+                openAiModel,
+                algorithmFilter,
+                similarityFilter
+            );
+
+            if (results.Count > 0)
+            {
+                Utils.PrintComparisonTable(results);
+            }
+
+            logger.LogInformation("\nClosing database connection...");
+            // MongoClient does not implement IAsyncDisposable; sync disposal is intentional
+            dbClient?.Cluster?.Dispose();
+            logger.LogInformation("Database connection closed");
+        }
+        catch (Exception ex)
+        {
+            logger.LogError(ex, "Application failed");
+            Environment.ExitCode = 1;
+        }
+    }
+}

ai/select-algorithm-dotnet/README.md

@@ -0,0 +1,124 @@
+# Select Algorithm - .NET
+
+Compare DiskANN, HNSW, and IVF vector index algorithms across COS, L2, and IP similarity metrics using Azure DocumentDB.
+
+## Prerequisites
+
+- .NET 9.0+ SDK
+- Azure DocumentDB cluster
+- Azure OpenAI resource with `text-embedding-3-small` deployment
+- **RBAC roles**:
+  - `Cognitive Services OpenAI User` on the Azure OpenAI resource
+  - DocumentDB `dbOwner` role on the target database
+
+## Setup
+
+1. Set required environment variables using `dotnet user-secrets` or export them directly:
+
+```bash
+# Using dotnet user-secrets (recommended for local development)
+dotnet user-secrets init
+dotnet user-secrets set "AZURE_OPENAI_EMBEDDING_ENDPOINT" "https://<RESOURCE-NAME>.openai.azure.com"
+dotnet user-secrets set "AZURE_OPENAI_EMBEDDING_MODEL" "text-embedding-3-small"
+dotnet user-secrets set "MONGO_CLUSTER_NAME" "<CLUSTER-NAME>"
+dotnet user-secrets set "AZURE_TENANT_ID" "<TENANT-ID>"
+
+# Or export as environment variables
+export AZURE_OPENAI_EMBEDDING_ENDPOINT="https://<RESOURCE-NAME>.openai.azure.com"
+export AZURE_OPENAI_EMBEDDING_MODEL="text-embedding-3-small"
+export MONGO_CLUSTER_NAME="<CLUSTER-NAME>"
+export AZURE_TENANT_ID="<TENANT-ID>"
+```
+
+2. Restore dependencies:
+
+```bash
+dotnet restore
+```
+
+## Usage
+
+### Compare all algorithms (default: COS similarity)
+
+```bash
+dotnet run
+```
+
+Set `ALGORITHM` and `SIMILARITY` environment variables to control which collections are queried:
+
+| ALGORITHM | SIMILARITY | Collections queried |
+|-----------|------------|---------------------|
+| `all`     | `COS`      | 3 (one per algorithm, COS) |
+| `all`     | `all`      | 9 (all combinations) |
+| `diskann` | `COS`      | 1 (hotels_diskann_cos) |
+| `diskann` | `all`      | 3 (diskann x all similarities) |
+
+## Architecture
+
+Creates collections named `hotels_{algorithm}_{similarity}` for each combination:
+
+| Algorithm | COS | L2 | IP |
+|-----------|-----|----|----|
+| DiskANN   | `hotels_diskann_cos` | `hotels_diskann_l2` | `hotels_diskann_ip` |
+| HNSW      | `hotels_hnsw_cos` | `hotels_hnsw_l2` | `hotels_hnsw_ip` |
+| IVF       | `hotels_ivf_cos` | `hotels_ivf_l2` | `hotels_ivf_ip` |
+
+Each collection gets its own vector index created via `RunCommandAsync` and data inserted via `InsertManyAsync`. The application runs vector search aggregation queries and prints a comparison table with latency metrics.
+
+## Algorithm-Specific Parameters
+
+### Index Creation
+- **DiskANN**: `maxDegree: 32`, `lBuild: 50`
+- **HNSW**: `m: 16`, `efConstruction: 64`
+- **IVF**: `numLists: 1`
+
+### Search Queries
+- **DiskANN**: `lSearch: 100`
+- **HNSW**: `efSearch: 80`
+- **IVF**: `nProbes: 1`
+
+## Authentication
+
+Uses `DefaultAzureCredential` for passwordless authentication to both Azure OpenAI and DocumentDB. Ensure you are logged in with Azure CLI:
+
+```bash
+az login
+```
+
+## Expected output
+
+Running with default settings (`ALGORITHM=all`, `SIMILARITY=COS`) prints a comparison table. Actual timings and scores vary per run.
+
+```
+==========================================================================================
+                     Vector Algorithm Comparison Results
+==========================================================================================
+Algorithm     Similarity    Top Result                Score         Latency(ms)
+------------------------------------------------------------------------------------------
+DiskANN       COS           Historic Downtown Inn      0.8342        45
+HNSW          COS           Historic Downtown Inn      0.8342        38
+IVF           COS           Historic Downtown Inn      0.8342        52
+==========================================================================================
+
+--- DiskANN / COS (hotels_diskann_cos) ---
+  1. Historic Downtown Inn              (Score: 0.8342)
+  2. Mountain Trail Lodge                (Score: 0.7891)
+  3. Riverside Retreat                   (Score: 0.7654)
+  4. Urban Fitness Suites                (Score: 0.7210)
+  5. Lakeside Wellness Resort            (Score: 0.7045)
+```
+
+> Note: Results vary based on data, embeddings, and server load. The table above is representative only.
+
+## Troubleshooting
+
+| Problem | Resolution |
+|---------|------------|
+| **OIDC authentication failure** | Verify `DefaultAzureCredential` configuration. Run `az login` (or `az login --tenant <TENANT-ID>`) and confirm the correct subscription is active. |
+| **Data file not found** | Verify the `Hotels_Vector.json` path. By default the app resolves `../../data/Hotels_Vector.json` relative to the build output directory. Override with the `DATA_FILE_WITH_VECTORS` environment variable. |
+| **Connection timeout** | Check network firewall rules and confirm the DocumentDB cluster is running and accessible. Ensure TLS is enabled and the cluster name is correct. |
+| **Azure OpenAI errors** | Verify the endpoint URL, confirm the `text-embedding-3-small` model deployment exists, and ensure your identity has the `Cognitive Services OpenAI User` RBAC role. |
+
+## Important Notes
+
+- **Collection cleanup**: This sample drops and recreates collections on every run. Any existing data in collections matching the `hotels_{algorithm}_{similarity}` naming pattern will be deleted.

ai/select-algorithm-dotnet/SelectAlgorithm.csproj

@@ -0,0 +1,34 @@
+<Project Sdk="Microsoft.NET.Sdk">
+
+  <PropertyGroup>
+    <OutputType>Exe</OutputType>
+    <TargetFramework>net9.0</TargetFramework>
+    <Nullable>enable</Nullable>
+    <ImplicitUsings>enable</ImplicitUsings>
+    <Description>Compare DiskANN, HNSW, and IVF vector index algorithms using Azure DocumentDB</Description>
+    <Authors>Microsoft Corporation</Authors>
+    <Company>Microsoft Corporation</Company>
+    <RootNamespace>SelectAlgorithm</RootNamespace>
+    <RestorePackagesWithLockFile>true</RestorePackagesWithLockFile>
+  </PropertyGroup>
+
+  <ItemGroup>
+    <PackageReference Include="Azure.AI.OpenAI" Version="2.1.0" />
+    <PackageReference Include="Azure.Identity" Version="1.17.1" />
+    <PackageReference Include="MongoDB.Driver" Version="3.0.0" />
+    <PackageReference Include="Microsoft.Extensions.Configuration" Version="9.0.0" />
+    <PackageReference Include="Microsoft.Extensions.Configuration.EnvironmentVariables" Version="9.0.0" />
+    <PackageReference Include="Microsoft.Extensions.Configuration.Json" Version="9.0.0" />
+    <PackageReference Include="Microsoft.Extensions.DependencyInjection" Version="9.0.0" />
+    <PackageReference Include="Microsoft.Extensions.Logging" Version="9.0.0" />
+    <PackageReference Include="Microsoft.Extensions.Logging.Console" Version="9.0.0" />
+
+  </ItemGroup>
+
+  <ItemGroup>
+    <Content Include="appsettings.json">
+      <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
+    </Content>
+  </ItemGroup>
+
+</Project>