1 of 13

Nimbus: Introduction

by various contributors�(see last slide)

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

2 of 13

Request for contributors

Hi and thanks for your contribution!

 

If you have modified something, I'm kindly asking you to

- put your name in the last slide

- write a quick note the next slide concerning what you have modified/added/deleted (since googledocs does not provide this feature, it only says who has modified the presentation and how many slides have been modified)

�

Massimo Canonico (mex@di.unipmn.it)

If you want to became an editor, just send me an email!

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

3 of 13

Revision history

�

01/12/11 - M. Canonico - sent the draft to cloud ML/forum to get feedback

18/11/11 - M. Canonico - first draft

17/11/11 - M. Canonico - uploaded on GD

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

4 of 13

ToDO list

  • Improve/Fix the layout
  • Are there more information to put in this introduction?

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

5 of 13

Outline

  • What is Nimbus?
  • Architecture
  • Components
  • Science Cloud: goals
  • Projects and Cloud Computing
  • Cloud Computing and projects
  • Nimbus on FutureGrid

​

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

6 of 13

What is Nimbus?

  • Set of open source tools that together provide an "Infrastructure-as-a-Service" (IaaS)

​

  • The programming framework used is Java and Python. The virtualization technologies supported are XEN and KVM.

​

​

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

7 of 13

What is Nimbus (II)?

  • An extensible open source Infrastructure-as-a-Service implementation
    • Turns your cluster into a cloud
  • Why open source IaaS?
    • Experiment and use: make your own cloud or configure a private cloud
    • Customize: try new things, make the IaaS paradigm work for your application domain
  • Particular interest in customization: scientific computing

​

​

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

8 of 13

Architecture

EC2 Client

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

9 of 13

Components

  • The Workspace Service is a standalone site VM manager that different remote protocol frontends can invoke.
    • A Web Services Resource Framework (WSRF) based remote protocol implementation
    • An EC2 based remote protocol implementation of their SOAP and Query APIs (partial) that supports EC2 Clients
  • Cumulus is an open source implementation of the Amazon S3 REST API. It is used as the Nimbus repository solution and can also be installed standalone.
  • The cloudclient aims to get users up and running in minutes with instance launches and one-click clusters.

​

​

​

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

10 of 13

Components (II)

​

  • The WorkspacePilot allows you to integrate VMs with resources already configured to manage jobs (i.e., already using a batch scheduler like PBS).
  • Resource manager is used for a pool of physical nodes. It deploys and manages Workspaces on the nodes
  • The workspace-control agent implements VMM and network specific tasks on each hypervisor.
  • The ContextBroker allows clients to coordinate large virtual cluster launches automatically and repeatably.
  • The ContextAgent lives on VMs and interacts with the Context Broker at VM boot.

​

​

​

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

11 of 13

Science Clouds: Goals

​

  • Make it easy for scientific projects to experiment with cloud computing
    • Can cloud computing be used for science?

​

  • Customize software in response to the needs of scientific projects
    • Start with EC2-like functionality and evolve to serve scientific projects: virtual clusters, diverse resource leases
    • Federating clouds: moving between cloud resources in academic and commercial space

​

​

​

​

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

12 of 13

Projects and Cloud Computing

​

  • CloudBLAST: bioinformatics applications
    • CS research: investigate latency-sensitive apps, e.g. hadoop
    • Need access to distributed resources, and high level of privilege to run a ViNE router
    • Virtual workspace: ViNE router + application VMs

​

  • STAR: a high-energy physics experiment
    • Needs resources with the right configuration
      • Complex environments: correct versions of operating systems, libraries, tools, etc all have to be installed.
      • Consistent environments: require validation

​

​

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it

13 of 13

Nimbus on FutureGrid

​

  • Hotel (University of Chicago) -- Xen
    • 41 nodes, 328 cores
  • Foxtrot (University of Florida) -- Xen
    • 26 nodes, 208 cores
  • Sierra (SDSC) -- Xen
    • 18 nodes, 144 cores
  • Alamo (TACC) -- KVM
    • 15 nodes, 120 cores

​

​

​

Ph. D. Massimo Canonico - University of Piemonte Orientale (Italy) - massimo.canonico@uniupo.it