Slides from a talk at the Common Solutions Group (CSG) meeting at UC Berkeley, January 2015.
This is the story of the Berkeley Desktop which was a key component enabling the success of Berkeley’s IT Shared Services through the introduction of automation, bringing stability and security to computer / endpoint management as a service.

Let’s start at the beginning…

Fall 2011

OMG Version 0.1
- Everything is broken or breaking all the time
- No time for staff to work on solutions
- “Owned” (compromised) machines
- “Standard image” doesn’t work on laptops
- “Standard image” too hard to change
- “Standard image” BSODs because of new hardware, drivers
- Laptop users told to buy a desktop with the “standard image” to RDP into

The Problem
Imaging a machine circa 2011: 4 hours of senior tech time.
- Varying hardware standards
- No significant automation
- Manual work; no checklists
- Frontline support team includes senior techs
- Frontline support diverts the Brents to support most important and/or loudest senior executives




Tackle the Things That Increase Costs
- Labor
- Productivity loss
- Change
- Variance
Finally… on the path to DevOps
- Culture
- Automation
- Monitoring
- Instrumentation
- Reporting
- Agile
Joint Administrative Computing Standard

Reimaging a workstation — legacy process


Reimaging a workstation — EEI process

Workflow Automation Tools

New Employee Workflow Automation

2014

Time to Manually Complete Tasks

2014 — Self-Service for Windows

& a big win for CSS, Depts and campus
Time for self service? 1 minute, no ticket.
Software install counts

Self-Service for OS X

Coming soon…

More Technical Components
- Zabbix
- Splunk
- Powershell / Sapien
- GitHub
- Vagrant
- Docker
- Jenkins CI
- HA-Proxy
- Idonethis.com

The Berkeley Desktop Ecosystem

