Tutorial. Introduction to cluster configuration and management

Synopsis:

Instructors: André Carneiro and Bruno Fagundes / Laboratório Nacional de Computação Científica

Language: Portuguese

Prerequisites: Experience with *nix environments (e.g.,Unix, Linux) and using command line interfaces

Time: The tutorial will be carried out in 2 sessions of 4 hours each.

Topics Summary

  • HPC clusters Overview
    • Types of clusters
    • Basic architecture of an HPC cluster
  • Infrastructure components
    • Resource management
    • Time sincronism
    • Image repository
    • Shared file system
    • User account managemen
    • Passwordless authentication
    • Environment modules
    • High performance network
  • Hands-on lab environment
    • Infrastructure
    • Infrastructure tools
  • clustershell
  • Environment Modules
  • SLURM
  • Archtecture
  • Daemons
  • Configuretion Files
    • Partitions
    • Nodes
    • Scheduling policies
  • Administrative commands
    • Environment deployment
  • Munge
  • MySQL/MariaDB
  • Daemons
  • Slurmf configuration
  • Testing
  • Access control and Accounting
    • Accounting hierarchy
    • Implementing access control
    • Creating, modifying, and removing associations
    • Restricting resource access
  • Administrative activities
    • Modifying jobs’ priorities
    • Nodes management
    • Resource reservation
    • Partition management
    • Accounting report

Requirements

The mininum requirements to deploy the hands-on lab evironment are:

  • Softwares
  • VirtualBox 6.1
  • Hardware
  • Dual-Core processor with virtualization support
  • 4 GB RAM
  • 7 GB storage space