Backend.AI API Documentation

Latest API version: v4.20190315

Backend.AI is a hassle-free backend for AI model development and deployment. It runs arbitrary user codes safely in resource-constrained environments, using Docker and our own sandbox wrapper.

It hosts various programming languages and runtimes, such as Python 2/3, R, PHP, C/C++, Java, Javascript, Julia, Octave, Haskell, Lua and NodeJS, as well as AI-oriented libraries such as TensorFlow, Keras, Caffe, and MXNet.

Table of Contents

Concepts

• Key Concepts
  • Manager and Agents
  • Kernels (aka Compute Sessions)
  • Cluster Networking
  • Databases
  • Configuration Management
  • Virtual Folders
• API Overview
  • API KeyPair Registration
  • Accessing Admin APIs
• FAQ

Cluster Installation

• Quickstart Guides
  • Install from Source
  • Install from Package (Enterprise Edition)
• Supplementary Guides
  • Install Docker
  • Manually install CUDA at on-premise GPU servers
  • Let Backend.AI to utilize GPUs
  • Install dependencies for building Python
  • Install pyenv
  • Install pyenv’s virtualenv plugin
  • Install Python via pyenv
  • Create a virtualenv using a specific Python version
  • Activate the virtualenv for the current shell
  • Activate the virtualenv when your shell goes into a directory
  • Install monitoring and logging tools
  • Prepare Database for Manager

Client SDK

• Client SDK Libraries and Tools

API Common Reference

• API and Document Conventions
• Authentication
• Rate Limiting
• JSON Object References

User API Reference

• Introduction
• Kernel Management
• Service Ports (aka Service Proxies)
• Code Execution (Streaming)
• Code Execution (Query Mode)
• Code Execution (Batch Mode)
• Event Monitoring
• Virtual Folders
• Resource Presets

Admin API Reference

• Introduction
• KeyPair Management
• Compute Session Monitoring
• Virtual Folder Management
• Statistics

Developer Manuals

• Development Setup
• Adding New REPL Kernels

Indices and tables

• Index

• Module Index

• Search Page