Overview
On campus, there are thousands of computer systems. In order to be able to properly track the hardware it supports, IT Services needs to move towards automated processes and software tools to the greatest extent possible. Physical inventories are inherently labor-intensive, and they often yield incomplete results.
Current State
At the moment, inventory of resources that are covered under Service Level Agreements with IT Services is done irregularly.
The software application that IT Services uses to manage resources, BigFix, has the ability to provide hardware and software details about computers. A registry key (for PCs) or plist file (for Macs) can be placed on workstations so that BigFix can identify workstations covered under various support organizations. For some groups on campus, the BigFix site admins have set up a custom field so that support groups can enter the Stanford Asset tags. This will allow BigFix to be used for inventory, since theoretically all supported machines have BigFix installed on them.
Vision
Automation and back-end tools are the trend in this area. Physical inventories of certain resources will not completely go away in the foreseeable future, but automation should be utilized to the greatest extent possible. Any solution must work on both the Mac and PC platforms, and our goal should be to be able to perform an inventory with minimum disruption to our clients. Any solution must be scalable from a handful of resources to many hundreds or even thousands of resources.
The more resources that need to be inventoried, the more important having an automated process becomes. So in cases where clients pay IT Services a per-computer fee for supporting a large number of computers, an automated process makes it easier to provide accurate numbers quickly. The need for inventory numbers is more pressing in the fall, when Service Level Agreements (SLAs) are signed between IT Services and its clients. Although the focus for inventory will be on larger clients, it's possible that any solution could be tested on a smaller client, if not on a subset of the larger client. And smaller clients would benefit from the automation that is necessary for larger clients.
Goals
- Explore ways to better utilize existing tools (e.g., BigFix, NetDB and DHCPLog searches for networkable devices).
- Explore scripting as a means of automating searches and/or combining data from various sources.
- Explore ways of inventorying resources that aren't currently tracked (e.g., PDAs, smartphones).
Roadmap
- Create a project that brings the appropriate people from IT Services together to examine existing tools, and what features are either being underutilized or not utilized at all.
- Determine how to utilize scripting to automatically pull information from BigFix.
- All the functionality is available today within BigFix to track various hardware inventory, and a meeting is needed to determine a common set of properties to be retrieved. Once the list is set, an analysis can be created that targets all support machines to be inventoried. The analyses can be modified over time as needs change. Automated reports or on-demand reports can be exported out from the BigFix Console or via Web Reports.
- Due to privacy concerns, BigFix will not be used to track information on student and privately-owned machines. The data retrieved will only be on supported Stanford machines with an approved console operator.
- Develop a process for distributing that information to departments covered under SLA with IT Services.
- Make results available to the broader community for machines not covered by SLA with IT Services.
Measures of success
- Inventory information can be retrieved easily and automatically.
- Independent verification that tools work as expected.
