Back to articles
AI & Cloud

Deploy Qdrant on VPS: Breakthrough Image Similarity Search for E-Commerce Apps

A detailed guide to deploying Qdrant, a high-performance vector database, on a VPS to build image similarity search for e-commerce applications, enhancing user experience and...

DPTCloud Engineering4 min read
Deploy Qdrant on VPS: Breakthrough Image Similarity Search for E-Commerce Apps

Qdrant is an open-source vector database designed for fast similarity search using vector embeddings. In e-commerce applications, image similarity search enables users to find related products by uploading an image, improving shopping experience and conversion rates.

Deploying Qdrant on a VPS allows businesses to maintain full control over infrastructure, manage data securely, and reduce cloud service costs. This guide walks you through the setup, configuration, integration with AI models, and performance testing.

Requirements and Environment Setup

To deploy Qdrant on a VPS, you need:

  • A VPS instance (minimum 4GB RAM, 2 vCPUs, 20GB SSD)
  • Operating system: Ubuntu 22.04 or Debian 11
  • Python 3.8+ and pip
  • Docker and Docker Compose (optional)
  • SSH access to the VPS

We recommend using High-RAM Cloud VPS for Self-Hosted LLMs & API Gateways to ensure performance for AI workloads.

Deploying Qdrant on VPS

You can deploy Qdrant using Docker or directly. Here's the Docker-based approach:

# Create directory for Qdrant
mkdir qdrant && cd qdrant

# Create docker-compose.yml
cat <<EOF > docker-compose.yml
version: '3.8'

services:
  qdrant:
    image: qdrant/qdrant:latest
    ports:
      - "6333:6333"
    volumes:
      - ./qdrant_data:/qdrant/data
    environment:
      - QDRANT__LOG_LEVEL=INFO
EOF

# Start Qdrant
docker-compose up -d

After running the above commands, Qdrant will be accessible on port 6333. Verify it with:

curl -X GET http://localhost:6333/health

The response should be {"status": "ok"}.

To perform image similarity search, you need:

  1. Extract embeddings from images using models like ResNet, CLIP, or ViT.
  2. Insert embeddings into Qdrant.
  3. Search for similar vectors when a user uploads an image.

Here's an example using Python and the qdrant-client library:

from qdrant_client import QdrantClient
from qdrant_client.models import PointStruct, VectorParams, Distance
import numpy as np

# Connect to Qdrant
client = QdrantClient(host="localhost", port=6333)

# Create collection
client.create_collection(
    collection_name="image_search",
    vectors_config=VectorParams(size=512, distance=Distance.COSINE)
)

# Insert vector (assuming image embedding is already extracted)
embedding = np.random.rand(512).tolist()
client.upsert(
    collection_name="image_search",
    points=[PointStruct(id=1, vector=embedding, payload={"image_id": "img_001"})]
)

# Search for similar vectors
results = client.search(
    collection_name="image_search",
    query_vector=embedding,
    limit=5
)
print(results)

For embedding extraction, you can use models like clip or resnet50 in PyTorch.

Testing and Optimization

To test performance:

  • Run multiple concurrent queries to measure latency.
  • Monitor system resources (CPU, RAM, I/O) on the VPS.
  • Adjust Qdrant configuration in config.yaml if needed.

For optimization, consider:

Troubleshooting and Resources

Common issues:

  • Cannot connect to Qdrant: Check port 6333, ensure Docker is running, and verify docker-compose.yml is correct.
  • Low RAM: Increase VPS RAM or reduce the number of vectors loaded.
  • Slow performance: Optimize embedding models or use GPU if available.

Reference resources:

Assumptions: This guide assumes basic knowledge of Docker, Python, and AI. It does not assume pre-trained embedding models.

DPTCloud Tier III Infrastructure

High-RAM Cloud VPS Plans Optimized for AI, Databases & Cache

Enterprise RAID 10 NVMe SSDs, CMC & FPT District 9 uplinks with Anycast Anti-DDoS and 99.99% SLA.

View 22 VPS Plans
ECO · vn-02Recommended

VPS-4E

297.000 ₫ / mo

10% VAT included

  • CPU: 4 vCPU
  • RAM: 8 GB
  • SSD: 50 GB NVMe
ECO · vn-02

VPS-5E

614.900 ₫ / mo

10% VAT included

  • CPU: 8 vCPU
  • RAM: 16 GB
  • SSD: 100 GB NVMe
ADVANCED · vn-03Recommended

VPS-4T

548.900 ₫ / mo

10% VAT included

  • CPU: 4 vCPU
  • RAM: 8 GB
  • SSD: 50 GB NVMe

Related Articles

View all
ZaloZalo SupportDiscord0384 753 478