History of Computing at Physical Plant
EARLY COMPUTING
In the late 1980s, Physical Plant realized it would benefit from computerizing its records, and some
of its processes. Early attempts at computing were strange: the first computer was rolled around
the department on a cart.
Eventually, CIS began purchasing computers and equipping managers and clerks with hand-built computers.
These were 486 model processors with far less computing power than your average cell phone. But they proved
extremely useful and computing expanded.
An early CIS goal was to create an environment in which the user workstations were interchangeable.
If a PC broke down, a replacement could be issued and repairs could be made, and the user would not suffer
much down time.
Originally, a workstation connected to the network using an IBM Token Ring short card and a 3½ inch floppy disk
containing the connectivity programs. That system evolved into a configuration where most PC's contained a network card
capable of Remote Program Loading (RPL). RPL is a method developed by IBM in the mid 1980's that allows workstations to
boot up and connect to a network without requiring a bootable device onboard.
RPL allows the network card to act as a primary boot device by taking over the boot floppy, then connecting to the network
on a very low level. The modern PXE boot method is similar to this technology.
The card "calls out" to the network in a language reserved for RPL, communicating the computer's presence. A network
server listening for those requests can issue an image of a boot disk (containing network connection programs) to the card.
The card uses that image to connect on a high level, then unhooks from the floppy.
The process is fast and efficient and has a major benefit of easy upgrade. Instead of having to configure each workstation, the
old disk image is changed on the file server so one change affects many.
The RPL process was abandoned around 1997 in favor of installing software on and booting from a local hard disk.
BENEFITS OF NETWORKED COMPUTERS
When Physical Plant CIS implemented a network in 1985, the emphasis was on a server-centric network. It was decided that
workstations would not contain hard drives, but would use only server hard drives for program, operating system, and data storage.
For a long time, most Physical Plant computers had no user-useable hard drive.
Although technology has changed greatly since its creation, Physical Plant CIS continues to maintain a highly efficient,
server-oriented system that leverages the network to provide nearly uninterruptable service.
Workstations with severe or difficult to correct problems can be swapped out quickly with little down time.
Upgrading is relatively quick. Since user files are stored on the network, they don't have to be copied onto the target machine.
This means dreaded "disk crashes" don't have the negative impact normally associated with them.
Because all Physical Plant user computers are connected to the same network, any user may log onto any computer, and experience
mostly the same computing experience (we say mostly, because some configurations like colors, themes, and profiles remain on the
C: drive).
Third party imaging software permits CIS to create and configure a base install image. This image includes the operating system, a wide
range of drivers, and basic applications software including the current office suite. Using a preconfigured image cuts setup time
to a minimum. 30-minute setup times are possible to workflow is kept moving.
All files stored on the CIS system are backed up. An extensive library of tapes ensures files from months ago are still retrievable.
In case of natural disaster, it is relatively easy to start over with new hardware, since all our valuable data is kept safe.
Taking a proactive stance on all these facets of computing helps keep CIS' support costs low.
|
