Lab
  Publications
  Research
    Active projects
        AIS
  Open Source
 

 Resources
  Security
  Conferences
  Group Messaging
  Scientific
 

AIS: Adaptable Installation System

Utility computing, clusters-on-demand, complex compute farms, and other dynamic large scale, multi-configuration environments have created a need for automated, fast, and resource efficient system installation and reinstallation. The current main approaches: image-based and meta-data based OS installs do not fully satisfy this need. Imaging is fast, but not scalable to a large number of system configurations. Meta-data based installs scale well, but are significantly slower than imaging. The Adaptable Installation System, AIS, combines the strengths of the two approaches and facilitates fast, disk space efficient, manageable and scalable installation for heterogeneous environments.

Latest News

05/06/2004 An updated version of an AIS report is posted in the "Papers/Documentation" section.
05/01/2004 AIS website goes live.
04/20/2004 AIS version 0.1 completed

Papers/Documentation

For a full understading of the system, read the first two papers. The readme document provides hands-on instructions of how to use AIS, while the report provides a top level overview of its operations.

  • AIS Readme. (5/1/2004)[pdf]
  • "AIS: A Fast, Disk Space Efficient ’Adaptable Installation System ’ Supporting Multitudes of Diverse Software Configurations." A technical report on AIS. (5/6/2004) [pdf] [doc]
  • "AIS - Fast, disk-efficient system installation for heterogeneous environments: How to improve on Imaging and Kickstart?" An extended abstract of a paper submitted to Usenix LISA 2004 conference. (4/20/2003) [pdf]

Souce Code

Latest version of AIS can be downloaded here. It is recommended that you read the above papers to get an idea of how AIS can help you best. CVS repository will be available soon.

Comments

Send any comments, suggestions to AIS developers.


Questions or comments to:
webmaster@ensl.cs.gwu.edu
TEL: (202) 994-xxxx
FAX: (202) 994-4875
Experimental Networked Systems Lab
Computer Science Department
George Washington University
707 22nd Street, NW, Suite 301
Washington, DC 20052