1982 First Hewlett-Packard Minicomputer Grant


The 1982 HP grant, by 2002ís standards, was a rather pedestrian system:  an HP3000 Series 44, a refrigerator size 1 MB RAM CPU unit with 2 120-MB disc packs (each the size of a two draw file cabinet).  There were 15 "dumb" terminals hardwired to the system and a modem pool with a half dozen modems (mostly so some faculty could dial in from their offices).   But, oh, what a powerful difference the system made in the development of the technological evolution of the School, with many interesting stories related to its planning, arrival, and implementation.  And, an even more important political story related to the entire nature of computing on campus.

Campus Computing Reaction

From a UCLA central campus Office of Academic Computing (OAC) perspective, the HP system was seen as a major threat to its established base.  The only systems on campus other than the IBM 3033 mainframe were minicomputer systems (almost all DEC equipment) acquired by individual researchers acquired with grant funds and used exclusively within their grant constraints.  The Management School was looking toward a general-purpose system that would be used by a vast array of faculty and students.  The HP minicomputer would be a direct competitor of the IBM mainframe.  The HP system would not only provide online interactive computing, but would enable GSM to reallocate the RUs provided for students use to those faculty who were major computational users on the mainframe and were constantly running short.

I recall very soon after my assuming the role of director that McLean and I met with OACís director, Bill Kehl.  McLean had suggested that rather than GSM using its grant dollar to acquire terminals to be attached to the HP system, that GSM install a set of modems.  Then the students could used the Decwriters and dial the HP instead of the IBM.  We had even offered to negotiate a price to cover the lost revenue.  Kehl was adamant in his response: if GSM acquired a HP system, OAC would close the North-Campus facility, greatly inconveniencing several hundred users.  As it turned out, when we did install the HP in spring 1982, OAC came in and physically hardwired the Decwriters at 300 baud to the IBM mainframe, having them sit idle, rather than allow GSM to make use of these systems.  In retrospect I can understand Kehlís position:  this was the tip of an iceberg, and he was attempting to stop it before it sank his ship.  Central mainframe computing was being threaten by distributed minicomputers, and with it there was the complete lost of control, revenue, and prestige of what was once a very hollowed operation .

Space and Power

With the arrival of the HP 3000 we remodeled space in the old WDPC.  The "space wars" with OAC were probably the most aggravating aspect of the entire situation.  The arguments were around who "owned" the space.  The conflict was simple:  GSMís computing capability were growing to meet local demand and there was a corresponding reduction in need for the North-facility as previously configured.  GSM wanted the space.  OAC didnít want to relinquish it.  Some compromises were made, and over the next several years space swapped: key punch machines space became labs, storage rooms became offices, etc.

Preparing the space for the arrival of our first computer was another challenge.  I remember when the electrician came in to follow the HP power specs, they pulled a new separate subpanel into the adjoining room.  We all laughed that they brought so much power into the area.  Of course, the last laugh was on us;  a few years later a second panel had to be installed.  But it was the emergency buttons that were the real laugh.  These are the big red button youíre to hit if there is an emergency to turn off the power.  Well, it got installed and then, a couple years later when we had to move the system to the a connecting room which was larger, the buttons turned out to be in the wrong place:  behind the door!  When you opened the door it would swing around and hit the red button and turn off the power!!!!  That only happen a couple times before the door stop was installed.  And it took a couple years more to get the button moved to an opposite wall!  At least with an existing raised floor, it made the wiring for the room very clean.

The air conditioning for that room was another challenge. The original WDPC had tons of power and its own independent air conditioning and blower system.  Our first HP was fine with the ambient air.  However, with the move to the next room, and the doubling of the amount of equipment, disc drives, communications equipment, controls, and the like, more air was required.  So, an independent self contain unit was installed.  Water piping and drains had to be moved into this previous office only area.  The blower was installed on the ceiling, which for some unbelievable amount of time, dripped continuously, and of course equipment ended up being located directly under the drip.

An interesting note on the A/C.  When we had the 1994 Northridge earthquake, the very first item to be fixed within the school was that air conditioning unit so that the HP system could be kept running 24/7.  The HP became the communications link for the entire school, so its centrality was recognized.

HP systems use and impact

Let me quote from the 1983 proposal to the HP Foundation:  "A significant increase in the number of users is expected in Fall Quarter 1983.  The entering class of 440 MBA and Executive MBA students will be new users on the system while only 110 graduating students will leave it.  As all students are not computer users we expect an increase of about 300 new usersÖ.the number of users on the HP3000 system will grow by 42% to 830 users."  In other words, in 1983, we achieved 100% of our students having hands-on experiences with the system.  The Managerial Computing course (Mgmt 404) became mandatory in Fall 1983, and continued as a core requirement for almost a decade.

There is a set of graphs included in the 1983 grant proposal describing system growth and usage in terms of connect time and CPU utilization.

Curriculum Evolution

Starting in Fall 1983, I taught an entirely new and revised Managerial Computing course.  I still remember my first two assignments clear:  each student was pretend they were seeking a job and log onto the HP to access a program which simulated an on-line application.  The data was then stored in a HP Image database.  For assignment two, the students were to pretend they were the personnel officer seeking a certain profile of employee and write a query (using the HP Query language) to select a set of potential candidates.  Oh the fun we all had struggling with the new system, the programs, and challenges.  Students were very excited about the prospects of getting so close to all this mysterious technology and participating at a level that those with non-technical or science oriented undergraduate degree could every image.  During that period a very large number of our MBA students entered the system analysis field, using the combination of their business and computing backgrounds to great advantage.

After a couple years of hands-on hardware focus, the faculty moved the Managerial Computing course to a theoretical offering, which by the end of the decade had once again returned to an optional rather than required core course status.

Return to Timeline

jason.frand@anderson.ucla.edu
November 1, 2002