cp

README.md

@@ -9,7 +9,8 @@ SEREACT is a secure API for storing, organizing, and retrieving images with adva
 - API key authentication
 - **Asynchronous image processing with Pub/Sub and Cloud Functions**
 - **AI-powered image embeddings using Google Cloud Vision API**
-- **Semantic search using vector similarity in Pinecone**
+- **Semantic search using vector similarity with Qdrant Vector Database**
+- **Self-hosted vector database on Google Compute Engine VM**
 - **Automatic retry mechanism for failed processing (up to 3 attempts)**
 - Metadata extraction and storage
 - Image processing capabilities
@@ -25,6 +26,8 @@ sereact/
   │   ├── cloud-function/        # **Cloud Function for image processing**
   │   ├── cloud-run/             # Google Cloud Run configuration
   │   └── terraform/             # Infrastructure as code
+  │       ├── vm.tf              # **Vector database VM configuration**
+  │       └── scripts/           # **VM installation scripts**
   ├── docs/                      # Documentation
   │   ├── api/                   # API documentation
   │   └── TESTING.md             # Comprehensive testing guide
@@ -41,7 +44,8 @@ sereact/
   │   ├── models/                # Database models
   │   ├── schemas/               # API request/response schemas
   │   ├── services/              # Business logic services
-  │   │   └── pubsub_service.py  # **Pub/Sub message publishing**
+  │   │   ├── pubsub_service.py  # **Pub/Sub message publishing**
+  │   │   └── vector_db.py       # **Qdrant vector database service**
   │   └── utils/                 # Utility functions
   ├── tests/                     # Test code
   │   ├── api/                   # API tests
@@ -79,9 +83,9 @@ sereact/
                                ▼
                         ┌─────────────┐         ┌─────────────┐
                         │             │         │             │
-                        │  Cloud      │         │  Pinecone   │
+                        │  Cloud      │         │  Qdrant     │
                         │  Vision API │────────▶│  Vector DB  │
-                        │             │         │             │
+                        │             │         │  (VM)       │
                         └─────────────┘         └─────────────┘
 ```
 
@@ -98,7 +102,7 @@ sereact/
    - Function updates image status to `"processing"`
    - **Function downloads image from Cloud Storage**
    - **Function calls Google Cloud Vision API to generate embeddings**
-   - **Embeddings are stored in Pinecone Vector Database**
+   - **Embeddings are stored in Qdrant Vector Database on dedicated VM**
    - **Firestore is updated with embedding info and status: "success"**
 
 ### 3. **Error Handling & Retry**:
@@ -108,7 +112,7 @@ sereact/
 
 ### 4. **Search Flow**:
    - Search queries processed by FastAPI backend
-   - Vector similarity search performed against Pinecone
+   - Vector similarity search performed against Qdrant VM
    - Results combined with metadata from Firestore
 
 ## Technology Stack
@@ -119,16 +123,44 @@ sereact/
 - **Google Pub/Sub** - Message queue for async processing
 - **Google Cloud Functions** - Serverless image processing
 - **Google Cloud Vision API** - AI-powered image analysis and embedding generation
-- **Pinecone** - Vector database for semantic search
+- **Qdrant** - Self-hosted vector database for semantic search (on Google Compute Engine VM)
+- **Google Compute Engine** - VM hosting for vector database
 - **Pydantic** - Data validation
 
+## **Vector Database Infrastructure**
+
+### **Qdrant Vector Database VM**
+
+The system includes a dedicated Google Compute Engine VM running Qdrant vector database:
+
+- **VM Specifications**: 2 vCPUs, 8GB RAM, 50GB disk (e2-standard-2)
+- **Operating System**: Ubuntu 22.04 LTS
+- **Vector Database**: Qdrant (latest version via Docker)
+- **Ports**: 6333 (HTTP API), 6334 (gRPC API)
+- **Features**:
+  - Automatic installation and configuration via startup script
+  - Daily automated backups
+  - Health monitoring
+  - Firewall configuration
+  - Optional static IP assignment
+  - API key authentication support
+
+### **Vector Database Features**
+
+- **High Performance**: Optimized for image vector similarity search
+- **Scalable**: Can handle millions of image vectors
+- **Persistent Storage**: Data persisted on VM disk with automated backups
+- **RESTful API**: Easy integration with Python client
+- **Cosine Similarity**: Optimized for image embedding comparisons
+- **Metadata Filtering**: Support for complex search filters
+
 ## Setup and Installation
 
 ### Prerequisites
 
 - Python 3.8+
-- Google Cloud account with Firestore, Storage, Pub/Sub, Cloud Functions, and Vision API enabled
+- Google Cloud account with Firestore, Storage, Pub/Sub, Cloud Functions, Compute Engine, and Vision API enabled
-- Pinecone account for vector database
+- Terraform (for infrastructure deployment)
 
 ### Installation
 
@@ -170,34 +202,79 @@ sereact/
    # Security
    API_KEY_SECRET=your-secret-key
    
-   # Vector database (Pinecone)
+   # Vector database (Qdrant)
-   VECTOR_DB_API_KEY=your-pinecone-api-key
+   QDRANT_HOST=your-vm-external-ip
-   VECTOR_DB_ENVIRONMENT=your-pinecone-environment
+   QDRANT_API_KEY=your-qdrant-api-key  # Optional
-   VECTOR_DB_INDEX_NAME=image-embeddings
    ```
 
-5. **Deploy Infrastructure** (Optional - for production):
+5. **Deploy Infrastructure** (Required for vector database):
    ```bash
-   # Deploy Pub/Sub infrastructure with Terraform
+   # Configure Terraform variables
    cd deployment/terraform
+   cp terraform.tfvars.example terraform.tfvars
+   # Edit terraform.tfvars with your values
+   
+   # Deploy infrastructure including vector database VM
    terraform init
    terraform plan
    terraform apply
    
-   # Deploy Cloud Function
+   # Note the output values for VM IP addresses
+   ```
+
+6. **Deploy Cloud Function** (Optional - for production):
+   ```bash
    cd ../cloud-function
    ./deploy.sh
    ```
 
-6. Run the application:
+7. Run the application:
    ```bash
    uvicorn main:app --reload
    ```
 
-7. Visit `http://localhost:8000/docs` in your browser to access the API documentation.
+8. Visit `http://localhost:8000/docs` in your browser to access the API documentation.
 
 ## **Deployment**
 
+### **Complete Infrastructure Deployment**
+
+Deploy the entire infrastructure including the vector database VM:
+
+```bash
+cd deployment/terraform
+
+# Configure your variables
+cp terraform.tfvars.example terraform.tfvars
+# Edit terraform.tfvars with your specific values:
+# - project_id: Your GCP project ID
+# - storage_bucket_name: Unique bucket name
+# - qdrant_api_key: Secure API key for Qdrant (optional)
+# - allowed_cidr_blocks: Your IP address/range for security
+# - use_static_ip: Set to true for production
+
+# Deploy infrastructure
+terraform init
+terraform plan
+terraform apply
+```
+
+This will create:
+- **Google Compute Engine VM with Qdrant vector database**
+- **Firewall rules for vector database access**
+- **Service accounts and IAM bindings**
+- **Pub/Sub topic and subscription with retry policy**
+- **Cloud Storage bucket**
+- **Firestore database**
+- **Cloud Run service**
+
+### **Vector Database VM Outputs**
+
+After deployment, Terraform will output:
+- `vector_db_vm_external_ip`: External IP address of the VM
+- `qdrant_http_endpoint`: HTTP API endpoint for Qdrant
+- `qdrant_grpc_endpoint`: gRPC API endpoint for Qdrant
+
 ### **Cloud Function Deployment**
 
 The image processing Cloud Function can be deployed using the provided script:
@@ -207,34 +284,59 @@ cd deployment/cloud-function
 
 # Set environment variables
 export GOOGLE_CLOUD_PROJECT=your-project-id
-export PINECONE_API_KEY=your-pinecone-api-key
+export QDRANT_HOST=your-vm-external-ip
-export PINECONE_ENVIRONMENT=your-pinecone-environment
+export QDRANT_API_KEY=your-qdrant-api-key
 
 # Deploy the function
 ./deploy.sh
 ```
 
-### **Infrastructure as Code**
+### **Vector Database Management**
 
-Use Terraform to deploy the complete infrastructure:
+#### **Accessing the Vector Database**
 
 ```bash
-cd deployment/terraform
+# SSH into the VM
+gcloud compute ssh sereact-vector-db --zone=us-central1-a
 
-# Initialize Terraform
+# Check Qdrant status
-terraform init
+sudo systemctl status qdrant
 
-# Review the deployment plan
+# View logs
-terraform plan
+sudo journalctl -u qdrant -f
 
-# Deploy infrastructure
+# Run health check
-terraform apply
+sudo /opt/qdrant/health_check.sh
+
+# Manual backup
+sudo /opt/qdrant/backup.sh
 ```
 
-This will create:
+#### **Vector Database API Usage**
-- **Pub/Sub topic and subscription with retry policy**
+
-- **Dead letter queue for failed messages**
+```python
-- **IAM bindings for service accounts**
+from src.services.vector_db import VectorDatabaseService
+
+# Initialize service
+vector_db = VectorDatabaseService(
+    host="your-vm-external-ip",
+    api_key="your-qdrant-api-key"  # Optional
+)
+
+# Add image vector
+point_id = vector_db.add_image_vector(
+    image_id="img_123",
+    vector=[0.1, 0.2, ...],  # 512-dimensional vector
+    metadata={"filename": "image.jpg", "size": 1024}
+)
+
+# Search similar images
+results = vector_db.search_similar_images(
+    query_vector=[0.1, 0.2, ...],
+    limit=10,
+    score_threshold=0.7
+)
+```
 
 ## API Endpoints
 
@@ -244,7 +346,7 @@ The API provides the following main endpoints:
 - `/api/v1/teams/*` - Team management
 - `/api/v1/users/*` - User management
 - `/api/v1/images/*` - **Image upload, download, and management (with async processing)**
-- `/api/v1/search/*` - **Image search functionality (semantic search)**
+- `/api/v1/search/*` - **Image search functionality (semantic search via Qdrant)**
 
 ### **Image Processing Status**
 
@@ -273,6 +375,7 @@ pytest
 
 # Run specific test categories
 pytest tests/services/test_pubsub_service.py  # Pub/Sub service tests
+pytest tests/services/test_vector_db.py       # Vector database tests
 pytest tests/integration/test_cloud_function.py  # Cloud Function tests
 pytest tests/api/test_images_pubsub.py        # API integration tests
 ```
@@ -292,6 +395,26 @@ python scripts/run_tests.py unit
 python scripts/run_tests.py integration
 ```
 
+## **Infrastructure Costs**
+
+### **Estimated Monthly Costs (USD)**
+
+- **Compute Engine VM (e2-standard-2)**: ~$50-70/month
+- **Cloud Storage**: $0.02/GB/month
+- **Firestore**: $0.18/100K reads, $0.18/100K writes
+- **Pub/Sub**: $0.40/million messages
+- **Cloud Functions**: $0.40/million invocations
+- **Cloud Vision API**: $1.50/1000 images
+
+**Total estimated cost for moderate usage**: ~$60-100/month
+
+### **Cost Optimization Tips**
+
+- Use preemptible VM instances for development (50-91% cost reduction)
+- Set up automatic VM shutdown during off-hours
+- Use regional persistent disks instead of SSD for cost savings
+- Monitor and set up billing alerts
+
 ## License
 
 This project is licensed under the MIT License - see the LICENSE file for details.
@@ -308,6 +431,7 @@ src/
   ├── config/          # Configuration management
   ├── models/          # Database models
   ├── services/        # Business logic services
+  │   └── vector_db.py # **Qdrant vector database service**
   └── utils/           # Utility functions
 ```
 
@@ -331,7 +455,7 @@ src/
 - Contains core business logic
 - Orchestrates operations across multiple resources
 - Implements domain-specific rules and workflows
-- Integrates with external services (Cloud Vision, Storage)
+- Integrates with external services (Cloud Vision, Storage, **Qdrant**)
 - Handles image processing and embedding generation
 
 #### Models Module
@@ -392,6 +516,7 @@ The modules interact in the following ways:
   - Auth Module validates the request authentication
   - Router delegates to appropriate Service functions
   - Service uses Models to interact with the database
+  - **Service integrates with Qdrant Vector Database for similarity search**
   - Service returns data to Router which formats the response
 
 - **Cross-Cutting Concerns**:

deployment/cloud-function/deploy.sh

@@ -13,9 +13,10 @@ MEMORY=${MEMORY:-"512MB"}
 TIMEOUT=${TIMEOUT:-"540s"}
 
 # Environment variables for the function
-PINECONE_API_KEY=${PINECONE_API_KEY:-""}
+QDRANT_HOST=${QDRANT_HOST:-""}
-PINECONE_ENVIRONMENT=${PINECONE_ENVIRONMENT:-""}
+QDRANT_PORT=${QDRANT_PORT:-"6333"}
-PINECONE_INDEX_NAME=${PINECONE_INDEX_NAME:-"image-embeddings"}
+QDRANT_API_KEY=${QDRANT_API_KEY:-""}
+QDRANT_COLLECTION=${QDRANT_COLLECTION:-"image_vectors"}
 
 echo "Deploying Cloud Function: $FUNCTION_NAME"
 echo "Project: $PROJECT_ID"
@@ -23,12 +24,9 @@ echo "Region: $REGION"
 echo "Pub/Sub Topic: $PUBSUB_TOPIC"
 
 # Check if required environment variables are set
-if [ -z "$PINECONE_API_KEY" ]; then
+if [ -z "$QDRANT_HOST" ]; then
-    echo "Warning: PINECONE_API_KEY not set. Function will not store embeddings."
+    echo "Warning: QDRANT_HOST not set. Function will not store embeddings."
-fi
+    echo "Please set QDRANT_HOST to your vector database VM's external IP address."
-
-if [ -z "$PINECONE_ENVIRONMENT" ]; then
-    echo "Warning: PINECONE_ENVIRONMENT not set. Function will not store embeddings."
 fi
 
 # Deploy the function
@@ -41,7 +39,7 @@ gcloud functions deploy $FUNCTION_NAME \
     --trigger-topic=$PUBSUB_TOPIC \
     --memory=$MEMORY \
     --timeout=$TIMEOUT \
-    --set-env-vars="PINECONE_API_KEY=$PINECONE_API_KEY,PINECONE_ENVIRONMENT=$PINECONE_ENVIRONMENT,PINECONE_INDEX_NAME=$PINECONE_INDEX_NAME" \
+    --set-env-vars="QDRANT_HOST=$QDRANT_HOST,QDRANT_PORT=$QDRANT_PORT,QDRANT_API_KEY=$QDRANT_API_KEY,QDRANT_COLLECTION=$QDRANT_COLLECTION" \
     --retry \
     --max-instances=10 \
     --min-instances=0
@@ -50,6 +48,7 @@ echo "Cloud Function deployed successfully!"
 echo "Function name: $FUNCTION_NAME"
 echo "Trigger: Pub/Sub topic '$PUBSUB_TOPIC'"
 echo "Region: $REGION"
+echo "Qdrant Host: $QDRANT_HOST"
 
 # Set up retry policy for the Pub/Sub subscription
 SUBSCRIPTION_NAME="${PUBSUB_TOPIC}-subscription"

deployment/cloud-function/main.py

@@ -7,11 +7,14 @@ import functions_framework
 from google.cloud import vision
 from google.cloud import firestore
 from google.cloud import storage
-import pinecone
+from qdrant_client import QdrantClient
+from qdrant_client.http import models
+from qdrant_client.http.models import Distance, VectorParams, PointStruct
 import numpy as np
 from PIL import Image
 import io
 import os
+import uuid
 
 # Configure logging
 logging.basicConfig(level=logging.INFO)
@@ -22,17 +25,47 @@ vision_client = vision.ImageAnnotatorClient()
 firestore_client = firestore.Client()
 storage_client = storage.Client()
 
-# Initialize Pinecone
+# Initialize Qdrant
-PINECONE_API_KEY = os.environ.get('PINECONE_API_KEY')
+QDRANT_HOST = os.environ.get('QDRANT_HOST', 'localhost')
-PINECONE_ENVIRONMENT = os.environ.get('PINECONE_ENVIRONMENT')
+QDRANT_PORT = int(os.environ.get('QDRANT_PORT', '6333'))
-PINECONE_INDEX_NAME = os.environ.get('PINECONE_INDEX_NAME', 'image-embeddings')
+QDRANT_API_KEY = os.environ.get('QDRANT_API_KEY')
+QDRANT_COLLECTION = os.environ.get('QDRANT_COLLECTION', 'image_vectors')
 
-if PINECONE_API_KEY and PINECONE_ENVIRONMENT:
+try:
-    pinecone.init(api_key=PINECONE_API_KEY, environment=PINECONE_ENVIRONMENT)
+    qdrant_client = QdrantClient(
-    index = pinecone.Index(PINECONE_INDEX_NAME)
+        host=QDRANT_HOST,
-else:
+        port=QDRANT_PORT,
-    index = None
+        api_key=QDRANT_API_KEY
-    logger.warning("Pinecone not configured, embeddings will not be stored")
+    )
+    
+    # Ensure collection exists
+    try:
+        collections = qdrant_client.get_collections()
+        collection_names = [col.name for col in collections.collections]
+        
+        if QDRANT_COLLECTION not in collection_names:
+            logger.info(f"Creating Qdrant collection: {QDRANT_COLLECTION}")
+            qdrant_client.create_collection(
+                collection_name=QDRANT_COLLECTION,
+                vectors_config=VectorParams(
+                    size=512,  # Fixed size for image embeddings
+                    distance=Distance.COSINE
+                )
+            )
+            logger.info(f"Collection {QDRANT_COLLECTION} created successfully")
+        else:
+            logger.info(f"Collection {QDRANT_COLLECTION} already exists")
+            
+    except Exception as e:
+        logger.error(f"Error ensuring Qdrant collection exists: {e}")
+        qdrant_client = None
+        
+except Exception as e:
+    logger.error(f"Failed to initialize Qdrant client: {e}")
+    qdrant_client = None
+
+if not qdrant_client:
+    logger.warning("Qdrant not configured, embeddings will not be stored")
 
 @functions_framework.cloud_event
 def process_image_embedding(cloud_event):
@@ -121,25 +154,36 @@ def process_image(image_id: str, storage_path: str, team_id: str, retry_count: i
             logger.error(f"Failed to generate embeddings for image {image_id}")
             return False
         
-        # Store embeddings in Pinecone
+        # Store embeddings in Qdrant
-        if index:
+        if qdrant_client:
-            embedding_id = f"{team_id}_{image_id}"
+            point_id = str(uuid.uuid4())
             
             # Prepare metadata
             metadata = {
                 'image_id': image_id,
                 'team_id': team_id,
                 'storage_path': storage_path,
-                'created_at': datetime.utcnow().isoformat()
+                'created_at': datetime.utcnow().isoformat(),
+                'model': 'google-vision-v1'
             }
             
-            # Upsert to Pinecone
+            # Create point for Qdrant
-            index.upsert(vectors=[(embedding_id, embeddings.tolist(), metadata)])
+            point = PointStruct(
+                id=point_id,
+                vector=embeddings.tolist(),
+                payload=metadata
+            )
             
-            logger.info(f"Stored embeddings for image {image_id} in Pinecone")
+            # Upsert to Qdrant
+            qdrant_client.upsert(
+                collection_name=QDRANT_COLLECTION,
+                points=[point]
+            )
+            
+            logger.info(f"Stored embeddings for image {image_id} in Qdrant with point ID {point_id}")
             
             # Update Firestore with embedding info
-            update_image_embedding_info(image_id, embedding_id, 'google-vision-v1')
+            update_image_embedding_info(image_id, point_id, 'google-vision-v1')
             
         return True
         
@@ -193,6 +237,39 @@ def generate_image_embeddings(image_data: bytes) -> Optional[np.ndarray]:
                 label.score  # Confidence score
             ])
         
+        # Get text detection for additional context
+        text_response = vision_client.text_detection(image=image)
+        
+        if text_response.text_annotations:
+            # Add text features
+            text_content = text_response.text_annotations[0].description if text_response.text_annotations else ""
+            text_hash = hash(text_content.lower()) % 1000 / 1000.0
+            features.extend([text_hash, len(text_content) / 1000.0])  # Normalized text length
+        
+        # Get face detection for additional features
+        face_response = vision_client.face_detection(image=image)
+        
+        face_count = len(face_response.face_annotations)
+        features.append(min(face_count / 10.0, 1.0))  # Normalized face count
+        
+        # Add image properties
+        try:
+            # Get image properties
+            properties_response = vision_client.image_properties(image=image)
+            
+            if properties_response.image_properties_annotation:
+                # Add dominant colors as features
+                colors = properties_response.image_properties_annotation.dominant_colors.colors
+                for i, color in enumerate(colors[:5]):  # Top 5 colors
+                    features.extend([
+                        color.color.red / 255.0,
+                        color.color.green / 255.0,
+                        color.color.blue / 255.0,
+                        color.score
+                    ])
+        except Exception as e:
+            logger.warning(f"Could not extract image properties: {e}")
+        
         # Pad or truncate to fixed size (512 dimensions)
         target_size = 512
         if len(features) < target_size:
@@ -200,7 +277,13 @@ def generate_image_embeddings(image_data: bytes) -> Optional[np.ndarray]:
         else:
             features = features[:target_size]
         
-        return np.array(features, dtype=np.float32)
+        # Normalize the feature vector
+        features_array = np.array(features, dtype=np.float32)
+        norm = np.linalg.norm(features_array)
+        if norm > 0:
+            features_array = features_array / norm
+        
+        return features_array
         
     except Exception as e:
         logger.error(f"Error generating embeddings: {e}")
@@ -238,20 +321,20 @@ def update_image_status(image_id: str, status: str, retry_count: int, error_mess
     except Exception as e:
         logger.error(f"Error updating image status: {e}")
 
-def update_image_embedding_info(image_id: str, embedding_id: str, model: str):
+def update_image_embedding_info(image_id: str, point_id: str, model: str):
     """
     Update the image with embedding information
     
     Args:
         image_id: The ID of the image
-        embedding_id: The ID of the embedding in the vector database
+        point_id: The ID of the point in the Qdrant vector database
         model: The model used to generate embeddings
     """
     try:
         doc_ref = firestore_client.collection('images').document(image_id)
         
         update_data = {
-            'embedding_id': embedding_id,
+            'embedding_point_id': point_id,
             'embedding_model': model,
             'has_embedding': True
         }

deployment/cloud-function/requirements.txt

@@ -2,6 +2,6 @@ functions-framework==3.4.0
 google-cloud-vision==3.4.5
 google-cloud-firestore==2.11.1
 google-cloud-storage==2.12.0
-pinecone-client==2.2.4
+qdrant-client==1.7.0
 numpy==1.24.3
 Pillow==10.1.0 

deployment/terraform/main.tf

@@ -11,7 +11,8 @@ resource "google_project_service" "services" {
     "containerregistry.googleapis.com",
     "run.googleapis.com",
     "firestore.googleapis.com",
-    "storage.googleapis.com"
+    "storage.googleapis.com",
+    "compute.googleapis.com"
   ])
   
   project = var.project_id

deployment/terraform/outputs.tf

@@ -17,3 +17,34 @@ output "container_registry_url" {
   value       = "gcr.io/${var.project_id}/sereact"
   description = "The URL of the Container Registry repository"
 }
+
+# Vector Database VM outputs
+output "vector_db_vm_name" {
+  value       = google_compute_instance.vector_db_vm.name
+  description = "The name of the vector database VM"
+}
+
+output "vector_db_vm_external_ip" {
+  value       = google_compute_instance.vector_db_vm.network_interface[0].access_config[0].nat_ip
+  description = "The external IP address of the vector database VM"
+}
+
+output "vector_db_vm_internal_ip" {
+  value       = google_compute_instance.vector_db_vm.network_interface[0].network_ip
+  description = "The internal IP address of the vector database VM"
+}
+
+output "vector_db_static_ip" {
+  value       = var.use_static_ip ? google_compute_address.vector_db_static_ip.address : null
+  description = "The static IP address of the vector database VM (if enabled)"
+}
+
+output "qdrant_http_endpoint" {
+  value       = "http://${google_compute_instance.vector_db_vm.network_interface[0].access_config[0].nat_ip}:6333"
+  description = "The HTTP endpoint for Qdrant vector database"
+}
+
+output "qdrant_grpc_endpoint" {
+  value       = "http://${google_compute_instance.vector_db_vm.network_interface[0].access_config[0].nat_ip}:6334"
+  description = "The gRPC endpoint for Qdrant vector database"
+} 

deployment/terraform/scripts/install_qdrant.sh

@@ -0,0 +1,189 @@
+#!/bin/bash
+
+# Qdrant Vector Database Installation Script
+# This script installs and configures Qdrant on Ubuntu 22.04
+
+set -e
+
+# Update system packages
+apt-get update
+apt-get upgrade -y
+
+# Install required packages
+apt-get install -y curl wget gnupg2 software-properties-common apt-transport-https ca-certificates
+
+# Install Docker
+curl -fsSL https://download.docker.com/linux/ubuntu/gpg | gpg --dearmor -o /usr/share/keyrings/docker-archive-keyring.gpg
+echo "deb [arch=amd64 signed-by=/usr/share/keyrings/docker-archive-keyring.gpg] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable" | tee /etc/apt/sources.list.d/docker.list > /dev/null
+apt-get update
+apt-get install -y docker-ce docker-ce-cli containerd.io
+
+# Start and enable Docker
+systemctl start docker
+systemctl enable docker
+
+# Create qdrant user and directories
+useradd -r -s /bin/false qdrant || true
+mkdir -p /opt/qdrant/storage
+mkdir -p /opt/qdrant/config
+chown -R qdrant:qdrant /opt/qdrant
+
+# Create Qdrant configuration file
+cat > /opt/qdrant/config/config.yaml << EOF
+service:
+  host: 0.0.0.0
+  http_port: 6333
+  grpc_port: 6334
+  enable_cors: true
+
+storage:
+  storage_path: /qdrant/storage
+  snapshots_path: /qdrant/snapshots
+  on_disk_payload: true
+
+cluster:
+  enabled: false
+
+telemetry:
+  disabled: true
+
+log_level: INFO
+EOF
+
+# Create API key configuration if provided
+if [ -n "${qdrant_api_key}" ] && [ "${qdrant_api_key}" != "" ]; then
+cat >> /opt/qdrant/config/config.yaml << EOF
+
+service:
+  api_key: "${qdrant_api_key}"
+EOF
+fi
+
+# Create systemd service for Qdrant
+cat > /etc/systemd/system/qdrant.service << EOF
+[Unit]
+Description=Qdrant Vector Database
+After=docker.service
+Requires=docker.service
+
+[Service]
+Type=simple
+User=root
+ExecStartPre=-/usr/bin/docker stop qdrant
+ExecStartPre=-/usr/bin/docker rm qdrant
+ExecStart=/usr/bin/docker run --name qdrant \
+  -p 6333:6333 \
+  -p 6334:6334 \
+  -v /opt/qdrant/storage:/qdrant/storage:z \
+  -v /opt/qdrant/config/config.yaml:/qdrant/config/production.yaml:z \
+  qdrant/qdrant:latest
+ExecStop=/usr/bin/docker stop qdrant
+Restart=always
+RestartSec=10
+
+[Install]
+WantedBy=multi-user.target
+EOF
+
+# Pull Qdrant Docker image
+docker pull qdrant/qdrant:latest
+
+# Enable and start Qdrant service
+systemctl daemon-reload
+systemctl enable qdrant
+systemctl start qdrant
+
+# Install monitoring tools
+apt-get install -y htop iotop nethogs
+
+# Create a simple health check script
+cat > /opt/qdrant/health_check.sh << 'EOF'
+#!/bin/bash
+response=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:6333/health)
+if [ "$response" = "200" ]; then
+    echo "Qdrant is healthy"
+    exit 0
+else
+    echo "Qdrant is not responding properly (HTTP $response)"
+    exit 1
+fi
+EOF
+
+chmod +x /opt/qdrant/health_check.sh
+
+# Set up log rotation for Docker logs
+cat > /etc/logrotate.d/docker << EOF
+/var/lib/docker/containers/*/*.log {
+    rotate 7
+    daily
+    compress
+    size=1M
+    missingok
+    delaycompress
+    copytruncate
+}
+EOF
+
+# Configure firewall (ufw)
+ufw --force enable
+ufw allow ssh
+ufw allow 6333/tcp  # Qdrant HTTP API
+ufw allow 6334/tcp  # Qdrant gRPC API
+
+# Create a simple backup script
+cat > /opt/qdrant/backup.sh << 'EOF'
+#!/bin/bash
+BACKUP_DIR="/opt/qdrant/backups"
+DATE=$(date +%Y%m%d_%H%M%S)
+mkdir -p $BACKUP_DIR
+
+# Create snapshot via API
+curl -X POST "http://localhost:6333/snapshots" \
+  -H "Content-Type: application/json" \
+  -d '{"snapshot_name": "backup_'$DATE'"}'
+
+# Copy storage directory
+tar -czf $BACKUP_DIR/qdrant_storage_$DATE.tar.gz -C /opt/qdrant storage/
+
+# Keep only last 7 backups
+find $BACKUP_DIR -name "*.tar.gz" -mtime +7 -delete
+
+echo "Backup completed: $DATE"
+EOF
+
+chmod +x /opt/qdrant/backup.sh
+
+# Set up daily backup cron job
+echo "0 2 * * * root /opt/qdrant/backup.sh >> /var/log/qdrant_backup.log 2>&1" >> /etc/crontab
+
+# Wait for Qdrant to be ready
+echo "Waiting for Qdrant to start..."
+for i in {1..30}; do
+    if curl -s http://localhost:6333/health > /dev/null; then
+        echo "Qdrant is ready!"
+        break
+    fi
+    echo "Waiting... ($i/30)"
+    sleep 10
+done
+
+# Create a default collection for image vectors
+curl -X PUT "http://localhost:6333/collections/image_vectors" \
+  -H "Content-Type: application/json" \
+  -d '{
+    "vectors": {
+      "size": 512,
+      "distance": "Cosine"
+    },
+    "optimizers_config": {
+      "default_segment_number": 2
+    },
+    "replication_factor": 1
+  }'
+
+echo "Qdrant installation and configuration completed!"
+echo "Qdrant is accessible at:"
+echo "  HTTP API: http://$(curl -s ifconfig.me):6333"
+echo "  gRPC API: http://$(curl -s ifconfig.me):6334"
+echo "Health check: /opt/qdrant/health_check.sh"
+echo "Backup script: /opt/qdrant/backup.sh" 

deployment/terraform/terraform.tfvars.example

@@ -3,3 +3,8 @@ region            = "us-central1"
 zone              = "us-central1-a"
 storage_bucket_name = "your-app-storage-bucket"
 firestore_db_name = "imagedb"
+
+# Vector Database Configuration
+qdrant_api_key = "your-secure-api-key-here"  # Optional: leave empty for no authentication
+allowed_cidr_blocks = "YOUR_IP_ADDRESS/32"   # Replace with your IP or network range
+use_static_ip = false                        # Set to true if you want a static IP 

deployment/terraform/variables.tf

@@ -49,3 +49,23 @@ variable "cloud_function_service_account" {
   type        = string
   default     = ""
 }
+
+# Vector Database VM variables
+variable "qdrant_api_key" {
+  description = "API key for Qdrant vector database"
+  type        = string
+  sensitive   = true
+  default     = ""
+}
+
+variable "allowed_cidr_blocks" {
+  description = "CIDR blocks allowed to access the vector database"
+  type        = string
+  default     = "0.0.0.0/0"  # Change this to your specific IP ranges for security
+}
+
+variable "use_static_ip" {
+  description = "Whether to use a static IP for the vector database VM"
+  type        = bool
+  default     = false
+} 

deployment/terraform/vm.tf

@@ -0,0 +1,104 @@
+# VM instance for vector database
+resource "google_compute_instance" "vector_db_vm" {
+  name         = "sereact-vector-db"
+  machine_type = "e2-standard-2"  # 2 vCPUs, 8GB RAM
+  zone         = var.zone
+
+  boot_disk {
+    initialize_params {
+      image = "ubuntu-os-cloud/ubuntu-2204-lts"
+      size  = 50  # 50GB disk
+      type  = "pd-standard"
+    }
+  }
+
+  network_interface {
+    network = "default"
+    access_config {
+      # Ephemeral public IP
+    }
+  }
+
+  # Startup script to install and configure Qdrant
+  metadata_startup_script = templatefile("${path.module}/scripts/install_qdrant.sh", {
+    qdrant_api_key = var.qdrant_api_key
+  })
+
+  # Service account for the VM
+  service_account {
+    email  = google_service_account.vector_db_sa.email
+    scopes = ["cloud-platform"]
+  }
+
+  # Tags for firewall rules
+  tags = ["vector-db", "qdrant"]
+
+  depends_on = [google_project_service.services]
+}
+
+# Service account for the vector DB VM
+resource "google_service_account" "vector_db_sa" {
+  account_id   = "vector-db-sa"
+  display_name = "Vector Database Service Account"
+  description  = "Service account for the vector database VM"
+}
+
+# Firewall rule to allow Qdrant access
+resource "google_compute_firewall" "qdrant_firewall" {
+  name    = "allow-qdrant"
+  network = "default"
+
+  allow {
+    protocol = "tcp"
+    ports    = ["6333", "6334"]  # Qdrant HTTP and gRPC ports
+  }
+
+  source_ranges = [
+    "10.0.0.0/8",     # Internal GCP networks
+    var.allowed_cidr_blocks  # Your specified IP ranges
+  ]
+
+  target_tags = ["qdrant"]
+}
+
+# Static IP for the vector DB VM (optional but recommended)
+resource "google_compute_address" "vector_db_static_ip" {
+  name   = "vector-db-static-ip"
+  region = var.region
+}
+
+# Attach the static IP to the VM
+resource "google_compute_instance" "vector_db_vm_with_static_ip" {
+  count        = var.use_static_ip ? 1 : 0
+  name         = "sereact-vector-db-static"
+  machine_type = "e2-standard-2"
+  zone         = var.zone
+
+  boot_disk {
+    initialize_params {
+      image = "ubuntu-os-cloud/ubuntu-2204-lts"
+      size  = 50
+      type  = "pd-standard"
+    }
+  }
+
+  network_interface {
+    network = "default"
+    access_config {
+      nat_ip = google_compute_address.vector_db_static_ip.address
+    }
+  }
+
+  metadata_startup_script = templatefile("${path.module}/scripts/install_qdrant.sh", {
+    qdrant_api_key = var.qdrant_api_key
+  })
+
+  service_account {
+    email  = google_service_account.vector_db_sa.email
+    scopes = ["cloud-platform"]
+  }
+
+  tags = ["vector-db", "qdrant"]
+
+  depends_on = [google_project_service.services]
+} 

requirements.txt

@@ -13,8 +13,8 @@ passlib==1.7.4
 tenacity==8.2.3
 pytest==7.4.3
 httpx==0.25.1
-pinecone-client==2.2.4
 pillow==10.1.0
 python-slugify==8.0.1 
 email-validator==2.1.0.post1
 pymongo==4.5.0
+qdrant-client==1.7.0

src/services/vector_db.py

@@ -0,0 +1,318 @@
+"""
+Vector Database Service for handling image vectors using Qdrant.
+"""
+
+import os
+import logging
+from typing import List, Dict, Any, Optional, Tuple
+import numpy as np
+from qdrant_client import QdrantClient
+from qdrant_client.http import models
+from qdrant_client.http.models import Distance, VectorParams, PointStruct
+import uuid
+
+logger = logging.getLogger(__name__)
+
+
+class VectorDatabaseService:
+    """Service for managing image vectors in Qdrant vector database."""
+    
+    def __init__(
+        self,
+        host: str = None,
+        port: int = 6333,
+        api_key: str = None,
+        collection_name: str = "image_vectors"
+    ):
+        """
+        Initialize the vector database service.
+        
+        Args:
+            host: Qdrant server host
+            port: Qdrant server port
+            api_key: API key for authentication (optional)
+            collection_name: Name of the collection to use
+        """
+        self.host = host or os.getenv("QDRANT_HOST", "localhost")
+        self.port = port
+        self.api_key = api_key or os.getenv("QDRANT_API_KEY")
+        self.collection_name = collection_name
+        
+        # Initialize Qdrant client
+        self.client = QdrantClient(
+            host=self.host,
+            port=self.port,
+            api_key=self.api_key
+        )
+        
+        # Ensure collection exists
+        self._ensure_collection_exists()
+    
+    def _ensure_collection_exists(self):
+        """Ensure the collection exists, create if it doesn't."""
+        try:
+            collections = self.client.get_collections()
+            collection_names = [col.name for col in collections.collections]
+            
+            if self.collection_name not in collection_names:
+                logger.info(f"Creating collection: {self.collection_name}")
+                self.client.create_collection(
+                    collection_name=self.collection_name,
+                    vectors_config=VectorParams(
+                        size=512,  # Typical size for image embeddings
+                        distance=Distance.COSINE
+                    )
+                )
+                logger.info(f"Collection {self.collection_name} created successfully")
+            else:
+                logger.info(f"Collection {self.collection_name} already exists")
+                
+        except Exception as e:
+            logger.error(f"Error ensuring collection exists: {e}")
+            raise
+    
+    def add_image_vector(
+        self,
+        image_id: str,
+        vector: List[float],
+        metadata: Dict[str, Any] = None
+    ) -> str:
+        """
+        Add an image vector to the database.
+        
+        Args:
+            image_id: Unique identifier for the image
+            vector: Image embedding vector
+            metadata: Additional metadata for the image
+            
+        Returns:
+            Point ID in the vector database
+        """
+        try:
+            point_id = str(uuid.uuid4())
+            payload = {
+                "image_id": image_id,
+                "timestamp": metadata.get("timestamp") if metadata else None,
+                "filename": metadata.get("filename") if metadata else None,
+                "size": metadata.get("size") if metadata else None,
+                "format": metadata.get("format") if metadata else None,
+                **(metadata or {})
+            }
+            
+            point = PointStruct(
+                id=point_id,
+                vector=vector,
+                payload=payload
+            )
+            
+            self.client.upsert(
+                collection_name=self.collection_name,
+                points=[point]
+            )
+            
+            logger.info(f"Added vector for image {image_id} with point ID {point_id}")
+            return point_id
+            
+        except Exception as e:
+            logger.error(f"Error adding image vector: {e}")
+            raise
+    
+    def search_similar_images(
+        self,
+        query_vector: List[float],
+        limit: int = 10,
+        score_threshold: float = 0.7,
+        filter_conditions: Dict[str, Any] = None
+    ) -> List[Dict[str, Any]]:
+        """
+        Search for similar images based on vector similarity.
+        
+        Args:
+            query_vector: Query vector to search for
+            limit: Maximum number of results to return
+            score_threshold: Minimum similarity score threshold
+            filter_conditions: Additional filter conditions
+            
+        Returns:
+            List of similar images with scores and metadata
+        """
+        try:
+            search_filter = None
+            if filter_conditions:
+                search_filter = models.Filter(
+                    must=[
+                        models.FieldCondition(
+                            key=key,
+                            match=models.MatchValue(value=value)
+                        )
+                        for key, value in filter_conditions.items()
+                    ]
+                )
+            
+            search_result = self.client.search(
+                collection_name=self.collection_name,
+                query_vector=query_vector,
+                query_filter=search_filter,
+                limit=limit,
+                score_threshold=score_threshold
+            )
+            
+            results = []
+            for hit in search_result:
+                result = {
+                    "point_id": hit.id,
+                    "score": hit.score,
+                    "image_id": hit.payload.get("image_id"),
+                    "metadata": hit.payload
+                }
+                results.append(result)
+            
+            logger.info(f"Found {len(results)} similar images")
+            return results
+            
+        except Exception as e:
+            logger.error(f"Error searching similar images: {e}")
+            raise
+    
+    def get_image_vector(self, image_id: str) -> Optional[Dict[str, Any]]:
+        """
+        Get vector and metadata for a specific image.
+        
+        Args:
+            image_id: Image identifier
+            
+        Returns:
+            Vector data and metadata if found, None otherwise
+        """
+        try:
+            search_result = self.client.scroll(
+                collection_name=self.collection_name,
+                scroll_filter=models.Filter(
+                    must=[
+                        models.FieldCondition(
+                            key="image_id",
+                            match=models.MatchValue(value=image_id)
+                        )
+                    ]
+                ),
+                limit=1,
+                with_vectors=True
+            )
+            
+            if search_result[0]:  # search_result is a tuple (points, next_page_offset)
+                point = search_result[0][0]
+                return {
+                    "point_id": point.id,
+                    "vector": point.vector,
+                    "image_id": point.payload.get("image_id"),
+                    "metadata": point.payload
+                }
+            
+            return None
+            
+        except Exception as e:
+            logger.error(f"Error getting image vector: {e}")
+            raise
+    
+    def delete_image_vector(self, image_id: str) -> bool:
+        """
+        Delete vector for a specific image.
+        
+        Args:
+            image_id: Image identifier
+            
+        Returns:
+            True if deleted successfully, False otherwise
+        """
+        try:
+            # First find the point ID
+            search_result = self.client.scroll(
+                collection_name=self.collection_name,
+                scroll_filter=models.Filter(
+                    must=[
+                        models.FieldCondition(
+                            key="image_id",
+                            match=models.MatchValue(value=image_id)
+                        )
+                    ]
+                ),
+                limit=1
+            )
+            
+            if search_result[0]:
+                point_id = search_result[0][0].id
+                self.client.delete(
+                    collection_name=self.collection_name,
+                    points_selector=models.PointIdsList(
+                        points=[point_id]
+                    )
+                )
+                logger.info(f"Deleted vector for image {image_id}")
+                return True
+            
+            logger.warning(f"No vector found for image {image_id}")
+            return False
+            
+        except Exception as e:
+            logger.error(f"Error deleting image vector: {e}")
+            raise
+    
+    def get_collection_info(self) -> Dict[str, Any]:
+        """
+        Get information about the collection.
+        
+        Returns:
+            Collection information including count and configuration
+        """
+        try:
+            collection_info = self.client.get_collection(self.collection_name)
+            return {
+                "name": collection_info.config.params.vectors.size,
+                "vectors_count": collection_info.points_count,
+                "vector_size": collection_info.config.params.vectors.size,
+                "distance": collection_info.config.params.vectors.distance,
+                "status": collection_info.status
+            }
+            
+        except Exception as e:
+            logger.error(f"Error getting collection info: {e}")
+            raise
+    
+    def health_check(self) -> bool:
+        """
+        Check if the vector database is healthy.
+        
+        Returns:
+            True if healthy, False otherwise
+        """
+        try:
+            collections = self.client.get_collections()
+            return True
+        except Exception as e:
+            logger.error(f"Vector database health check failed: {e}")
+            return False
+
+
+# Utility functions for vector operations
+def normalize_vector(vector: List[float]) -> List[float]:
+    """Normalize a vector to unit length."""
+    vector_array = np.array(vector)
+    norm = np.linalg.norm(vector_array)
+    if norm == 0:
+        return vector
+    return (vector_array / norm).tolist()
+
+
+def cosine_similarity(vector1: List[float], vector2: List[float]) -> float:
+    """Calculate cosine similarity between two vectors."""
+    v1 = np.array(vector1)
+    v2 = np.array(vector2)
+    
+    dot_product = np.dot(v1, v2)
+    norm_v1 = np.linalg.norm(v1)
+    norm_v2 = np.linalg.norm(v2)
+    
+    if norm_v1 == 0 or norm_v2 == 0:
+        return 0.0
+    
+    return dot_product / (norm_v1 * norm_v2)