Skip to content

Draft: GIS.lab workshop

Ludka Furtkevicova edited this page Apr 27, 2016 · 2 revisions

GIS.lab GOALS

  • rapid deployment of complete geospatial solution for collaborative data capturing, processing, analysis and publication on web

    • fully automatic provisioning
    • out-of-box working deployment with GIS.lab Unit
  • self containing system

  • very quick results (from plain hardware to web application in minutes)

  • high added values

    • integration of precisely chosen set of geospatial FOSS (one best tool for one thing) to one system with consistent behaviour - opposite to OSGeo Live
    • collaboration tools
    • user support, software support
  • full client computer performance utilisation (opposite to thin client)

  • thinking about client environment more as about some kind of specialized client interface providing tools from desktop world rather than a traditional desktop

  • computer resources sharing

  • same deployment in LAN and cloud

  • mobile clients (TODO)

  • web administration similar to router or NAS (TODO)

  • utilization: education and real work

HOW IS IT DONE ?

  • automatic provisioning - Ansible
  • virtual machine deployment - Vagrant and VirtualBox
  • client boot service - LTSP Fat client/ own solution
  • OWS services load balancing
  • QGIS Desktop and Server as GIS.lab Desktop + own GIS.lab Web app
  • GRASS as processing backend under QGIS Processing plugin and WPS
  • own GIS software packaging

GOALS OF THIS WORKSHOP

  • to provide technical knowledge required to build GIS.lab from source code and start using and hacking
  • get serious feedback to know what to change and improve
  • discuss GRASS 7 integration
  • discuss improvements needed for using GIS.lab in education
  • propose future development
  • get new contributors

WORKSHOP TOPICS

  • introduction to GIS.lab

  • introduction to Vagrant and Vagrant commands (status, up, destroy, provision, ssh)

  • GIS.lab installation in VirtualBox using Vagrant provisioner, configuration, update

  • introduction to GIS.lab management commands (adduser, deluser, password, machine)

  • extensibility with plugins and Docker containers

  • creation of GIS.lab network from scratch

  • booting physical client

  • booting virtual clients (HTTP boot, file sharing with host machine)

  • introduction to GIS.lab client environment (shared folders, Projects, Publish, Booster, software, services, PostgreSQL schemas, chat)

  • introduction to GIS.lab OWS services load balancing

  • creation of GIS projects in GIS.lab, publishing in GIS.lab Web

  • testing GRASS in GIS.lab

  • introduction to Ansible

  • introduction to GIS.lab source code

  • discussion about integration of education support tools

  • discussion about integration of GRASS 7

  • general discussion