home | sitemap | abstract | introduction | chaos | thinking | checklist | migrating | recovery pushpull | cost | career | workshop | isconf | list_and_community | papers | references |
Conclusion
There are many other ways this work could have been done, and many inconsistencies in the way we did things. One fact that astute readers will spot, for instance, is the way we used both file replication and a makefile to enact changes on client disks. While this rarely caused problems in practice, the most appropriate use of these two functions could stand to be more clearly defined. We welcome any and all feedback. This is the paper we wish we could have read many years ago. We hope that by passing along this information we've aided someone, somewhere, years in the future. If you are interested in providing feedback on this paper and helping improve the state of the art, we'd like to welcome you to our web site: Updates to this paper as well as code and contributions from others will be available at www.infrastructures.org. |
|
||||||||
© Copyright 1994-2007 Steve
Traugott,
Joel Huddleston,
Joyce Cao Traugott
In partnership with TerraLuna, LLC and CD International |