Backend Benchmark Repository

Contribution Guide

The Contribution Guide provides detailed instructions on:

• Adding new platforms to the benchmarks.
• Rerunning tests to validate results.
• Updating configurations for existing benchmarks.

Demo

Explore the interactive demo at backend-benchmark to:

• Compare different backend services.
• Visualize performance metrics for each framework and individual fields.

---

Table of Contents

• Introduction
• Prerequisites
• Completed Benchmarks
  • Python
  • Dart
  • JavaScript/TypeScript
  • C#
  • Go
  • Rust
• Testing Tool: Locust
• Benchmark Visualization
• Database Endpoints
• Static Endpoints

Introduction

This repository serves as a comprehensive resource for comparing backend technologies based on speed and load capabilities. Our benchmarks focus on write and read operations, as well as requests to static endpoints, reflecting real-world usage scenarios to assist developers and decision-makers in choosing the most suitable backend framework for their needs.

Prerequisites

• Docker installed on the system to run the benchmarks.
• The scripts/start_tests.sh script is used to launch each test and create the graphs.

Completed Benchmarks

Benchmarks are categorized into:

1. Database Tests (db_test): Involving database operations such as read and write requests.
2. Static Endpoint Tests (no_db_test): Involving requests to static endpoints without database interaction.

Python

• Django (Sync and Async) - Connection Pooling with PgBouncer.

Dart

• Serverpod

JavaScript/TypeScript

• Express (Node and Bun)

C#

• .Net Core

Go

Rust

Testing Tool: Locust

• Configuration:
  • Users: 10000
  • Spawn Rate: 83.33 users/second
  • Test Duration: 120 seconds

Benchmark Visualization

All benchmark results are visualized in the interactive dashboard:

View Live Dashboard

The dashboard includes:

• Rankings — Leaderboard with bar charts and sortable data table
• Detail — Single framework deep-dive with time series and percentile charts
• Compare — Head-to-head radar charts and grouped metric comparisons
• Time Series — Multi-framework line charts over test duration