Skip to main content

​
Welcome to Walle

Walle is a high-performance AVL (Automatic Vehicle Location) telemetry tracking API service built with NestJS. It provides robust infrastructure for collecting, storing, and querying vehicle GPS tracking data at scale using time-series partitioning and geospatial indexing.

Quickstart

Get Walle up and running in minutes with our step-by-step guide

Architecture

Understand Walle’s multi-database architecture and design patterns

API Reference

Explore the complete API documentation and data models

Partition Management

Learn how Walle manages time-series data with automated partitioning

​
Key Features

Walle is designed to handle high-volume vehicle telemetry data with enterprise-grade reliability:

​
Time-Series Partitioned PostgreSQL

Automatic daily partitioning for high-volume telemetry data ensures optimal query performance and efficient data retention management. Partitions are created automatically by the PartitionManagerService with scheduled jobs that run at 23:50 daily. 
// Automatic partition creation in main.ts:14
await partitionManager.createPartitionForDate(new Date());
await partitionManager.createPartitionForDate(new Date(Date.now() + 86_400_000));

​
JWT-Based Authentication

Secure authentication using Passport JWT strategy with 8-hour token expiration. The authentication module integrates with MongoDB for user management. 
// JWT configuration from auth.module.ts:16
signOptions: {
  expiresIn: '8h'
}

​
Multi-Database Architecture

Strategic separation of concerns with PostgreSQL for time-series telemetry data and MongoDB for user authentication and dispatch data:
  • PostgreSQL (walledb): Time-series partitioned storage for vehicle tracking points with PostGIS geospatial support
  • MongoDB (deltaDispatch): User and dispatch data management
  • MongoDB (authSoftware): Authentication and authorization data

​
Automated Partition Management

Scheduled cron jobs automatically create tomorrow’s partition at 23:50 daily, ensuring uninterrupted data collection: 
// From partition-manager.service.ts:111
@Cron('50 23 * * *')
async scheduleNextDay(): Promise<void> {
  const tomorrow = new Date();
  tomorrow.setDate(tomorrow.getDate() + 1);
  await this.createPartitionForDate(tomorrow);
}

​
Comprehensive Vehicle Tracking Data Model

The Point entity includes 50+ attributes for complete vehicle telemetry tracking:
  • Location data: Latitude, longitude, altitude, PostGIS geometry with SRID 4326
  • Movement metrics: Speed (m/s and km/h), angle, distance, movement state
  • Engine data: Ignition state, engine hours, fuel consumption
  • GNSS information: Satellite count, PDOP, HDOP, accuracy
  • Geofencing: Sector detection, quadrant tracking, jurisdiction zones
  • Hardware telemetry: GSM signal, battery level, network type
The Point model uses a composite primary key (id, timestamp) required for PostgreSQL range partitioning.

​
PostGIS Geospatial Indexing

GIST indexes on location columns enable efficient spatial queries for geofencing, proximity searches, and route analysis: 
-- From partition-manager.service.ts:82
CREATE INDEX idx_points_YYYYMMDD_location
  ON points_YYYYMMDD USING GIST (location)

​
Architecture Overview

Walle follows a modular NestJS architecture with clear separation of concerns: 
src/
├── app.module.ts              # Root module with database connections
├── main.ts                    # Bootstrap with partition initialization
└── app/
    ├── auth/                  # JWT authentication module
    │   ├── auth.module.ts
    │   └── strategies/
    │       └── jwt.strategy.ts
    ├── user/                  # User management (MongoDB)
    │   ├── user.module.ts
    │   ├── user.service.ts
    │   └── schema/user.schema.ts
    ├── points/                # Vehicle tracking points
    │   ├── points.module.ts
    │   ├── points.service.ts
    │   ├── points.controller.ts
    │   ├── partition-manager.service.ts
    │   └── model/point.model.ts
    └── database/              # Database configuration
        ├── database.module.ts
        ├── config/database-postgresql.config.ts
        └── constants/points.sql.ts

​
Use Cases

Walle is ideal for:
  • Fleet Management: Track vehicle locations, routes, and operational metrics in real-time
  • Public Safety: Monitor emergency vehicle positions and response times
  • Logistics: Analyze delivery routes, fuel efficiency, and driver behavior
  • Urban Planning: Collect traffic pattern data for infrastructure decisions
  • Asset Tracking: Monitor high-value mobile assets across large geographic areas

​
Next Steps

Get Started

Follow the quickstart guide to install and run Walle locally

Explore the API

Learn about the Point data model and available fields

Build docs developers (and LLMs) love

Get started for freeTalk to us