GC3: Grid Computing Competence Center

Presenting GC3 cloud system "Hobbes"

Who

This introductory seminar is aimed at those who want to start testing the GC3 Hobbes cloud infrastructure. More specifically this seminar is targeting those users who wants to implement their own scientific usecase levering the services that the GC3 group offers on its could infrastructure Hobbes.

Prerequisites

The seminar will cover all the basic generic aspects of cloud computing with a focus on how it can be empowered for scientific computing.

Familiarity with concepts like Virtual Machines management, basic Linux operating system as well as previous experiences with any cloud infrastructure will be an advantage.

For the scientific usecases part, we assume that you are already familiar with batch processing and have basic skills in either computer programming or in scripting languages (e.g. bash, perl, python, etc.)

What

Hobbes is an Infrastructure-as-a-Service solution specifically targeted to address large scale computational research, based on the OpenStack cloud management software.

Hobbes complements and completes the traditional computational clusters, and allows researchers to experiment with new usage models. Thanks to virtualization technologies, cloud systems allow building fully customized processing infrastructures, from a one-off experimentation with Matlab to continuous processing of laboratory data.

This small seminar covers the basics of Hobbes from the infrastructure point of view and the basic programming approaches that can be taken when implementing a large-scale scientific usecase on a similar type of infrastructure.

When

Feb 12 (Tuesday) 2013.

The seminar will start at 14:00 am and finish by 17:00.

Where

The training will take place at the UZH Irchel campus, in room Y11 F 06 (that's room 06 on floor F in building Y11).

Agenda

Download slides (pdf)

  • Preamble
    • a bit of terminology
  • Presenting Hobbes
    • online presentation
    • Demo 1: How to use ssh public key to access running instances (mp4)
    • Demo 2: How to create your own instance (mp4)
  • Enabling scientific usecases on cloud infrastructure
    • Few examples will be given to show and demonstrate how cloud technology could be lever for large scale data processing
    • Usecase presented by Stephan Gruber (GEO/UZH)
    • Demo 1: Simple scientiic usecase (mp4)
    • Demo 2: Using GC3Pie (mp4)
  • Bring your own usecase
    • We would like to hear from you what are your own usecases and discuss how they could be enabled on Hobbes

Other info

For further information, please write to: info@gc3.lists.uzh.ch

top