The first paper, presented by Freddie Chow of the Stanford Linear Accelerator Center (SLAC) described his teams efforts to create a support model which would scale with their projected growth. They established site-wide policies which were then executed by departmental system administrators. For NT, they deployed SMS to manage the systems.

The second paper, “Effective Usage of Individual User NT Profiles with Software Distribution” was presented by Mark Murray and Troy Roberts of Interprovincial Pipe Line Inc. (IPL). They created a home-grown system to manage their user accounts and their machines. The work occupied 12 developers for 18 months.

Remy Evard presented the final paper on building NT build clusters. His experiences with creating these clusters demonstrated some of the real limitations of large scale NT deployments. Most of the remote access products on the market won’t scale to 1000+ machine environments. More needs to be done to make NT useful from the command line (but more on that a little later).

Return to the index page.

Page created by Phil Scarr.