Skip to main content

Specialized PostgreSQL Variants

Appliku offers three specialized PostgreSQL builds that extend the standard engine with domain-specific capabilities. These are provisioned using the same workflow as regular PostgreSQL -- just select the specialized version from the dropdown.

Available Variants​

VariantBase VersionUse Case
PostGIS 16 + 3.4PostgreSQL 16Geospatial data and location queries
pgvectorPostgreSQL 16AI/ML vector similarity search
TimescaleDBPostgreSQL 15Time-series data and analytics

Provisioning​

The provisioning flow is identical to standard PostgreSQL:

  1. Open your application in the Appliku dashboard
  2. On the Application Overview page, scroll to the Databases block (bottom-right sidebar)
  3. Click Add Database
  4. Select the specialized variant from the dropdown (e.g., PostGIS 16+3.4)
  5. Select the server where the database should run
  6. Click Create

Connection URLs are automatically injected as environment variables, just like standard PostgreSQL.

PostGIS​

PostGIS adds geospatial data types, spatial indexing, and geographic query functions to PostgreSQL. It is the industry standard for storing and querying location data.

Version: PostgreSQL 16 + PostGIS 3.4

When to Use PostGIS​

  • Store and query geographic coordinates (points, polygons, lines)
  • Calculate distances between locations
  • Find nearby points of interest
  • Work with GeoJSON data
  • Build mapping or logistics applications

Django with PostGIS​

Django's django.contrib.gis (GeoDjango) has native PostGIS support:

DATABASES = {
    "default": {
        "ENGINE": "django.contrib.gis.db.backends.postgis",
        "NAME": "...",
        # Or parse from DATABASE_URL using dj-database-url with
        # the GIS engine override
    }
}

Ensure GDAL and GEOS libraries are available in your build image or Dockerfile.

pgvector​

pgvector adds vector data types and similarity search operators to PostgreSQL. It enables storing and querying high-dimensional embeddings directly in your database.

When to Use pgvector​

  • Store embeddings from OpenAI, Cohere, or other AI models
  • Perform nearest-neighbor similarity search
  • Build semantic search, recommendation engines, or RAG pipelines
  • Avoid running a separate vector database (e.g., Pinecone, Weaviate)

Example Usage​

-- Enable the extension
CREATE EXTENSION IF NOT EXISTS vector;

-- Create a table with a vector column
CREATE TABLE documents (
    id SERIAL PRIMARY KEY,
    content TEXT,
    embedding vector(1536)  -- OpenAI ada-002 dimension
);

-- Insert a vector
INSERT INTO documents (content, embedding)
VALUES ('Hello world', '[0.1, 0.2, ...]');

-- Find nearest neighbors
SELECT id, content
FROM documents
ORDER BY embedding <-> '[0.1, 0.2, ...]'
LIMIT 5;

Python with pgvector​

# pip install pgvector
from pgvector.psycopg2 import register_vector
import psycopg2

conn = psycopg2.connect(os.environ["DATABASE_URL"])
register_vector(conn)

TimescaleDB​

TimescaleDB extends PostgreSQL with hypertables optimized for time-series data. It provides automatic partitioning, compression, and time-based query optimizations.

Version: Based on PostgreSQL 15

When to Use TimescaleDB​

  • IoT sensor data and device telemetry
  • Application metrics and monitoring data
  • Financial time-series (stock prices, transactions)
  • Event logging with time-based queries
  • Any workload where data is primarily appended and queried by time range

Example Usage​

-- Create a regular table
CREATE TABLE metrics (
    time        TIMESTAMPTZ NOT NULL,
    device_id   INTEGER,
    temperature DOUBLE PRECISION,
    humidity    DOUBLE PRECISION
);

-- Convert to a hypertable
SELECT create_hypertable('metrics', 'time');

-- Insert data normally
INSERT INTO metrics (time, device_id, temperature, humidity)
VALUES (NOW(), 1, 22.5, 45.0);

-- Time-based queries are automatically optimized
SELECT time_bucket('1 hour', time) AS hour,
       AVG(temperature) AS avg_temp
FROM metrics
WHERE time > NOW() - INTERVAL '24 hours'
GROUP BY hour
ORDER BY hour;

Next Steps​