Control Data Corporation

Control Data Corporation (CDC) was a supercomputer firm. For most of the 1960s, it built the fastest computers in the world by far, only losing that crown in the 1970s after Seymour Cray left the company to found Cray Research, Inc. (CRI). CDC was one of the nine major United States computer companies through most of the 1960s; the others were IBM, Burroughs Corporation, DEC, NCR, General Electric, Honeywell, RCA, and UNIVAC. CDC was well known and highly regarded throughout the industry at one time.

Background and origins: World War II–1957

During World War II the U.S. Navy had built up a team of engineers to build codebreaking machinery for both Japanese and German electro-mechanical ciphers. A number of these were produced by a team dedicated to the task working in the Washington, D.C., area. With the post-war wind-down of military spending the Navy grew increasingly worried that this team would break up and scatter into various companies, and it started looking for ways to covertly keep the team together.

Eventually they found their solution; the owner of a Chase Aircraft affiliate in St. Paul, Minnesota, John Parker, was about to lose all his contracts with the end of the war. The Navy never told Parker exactly what the team did, since it would have taken too long to get top secret clearance. Parker was obviously wary, but after several meetings with increasingly high-ranking Naval officers it became apparent that whatever it was, they were serious, and he eventually agreed to give this team a home in his military glider factory.

The result was Engineering Research Associates (ERA), a contract engineering company that worked on a number of seemingly unrelated projects in the early 1950s. One of these was one of the first commercial stored program computers, the 36-bit ERA 1103. The machine was built for the Navy, which intended to use it in their "above board" code-breaking centers. In the early 1950s a minor political debate broke out in Congress about the Navy essentially "owning" ERA, and the ensuing debates and legal wrangling left the company drained of both capital and spirit. In 1952 Parker sold ERA to Remington Rand.

Although Rand kept the ERA team together and developing new products, it was most interested in ERA's magnetic drum memory systems. Rand soon merged with Sperry Corporation to become Sperry Rand, and in the process of merging the companies, the ERA division was folded into Sperry's UNIVAC division. At first this did not cause too many changes at ERA, since the company was used primarily to provide engineering talent to support a variety of projects. However, one major project was moved from UNIVAC to ERA, the UNIVAC II project, which led to lengthy delays and upsets to nearly everyone involved.

Since the Sperry "big company" mentality encroached on the decision-making powers of the ERA founders, they left Sperry to form the Control Data Corp. in 1957, setting up shop in an old warehouse down the road from Sperry in Minneapolis at 501 Park Avenue. Of the members forming CDC, William Norris was the unanimous choice to become the chief executive officer of the new company. Seymour Cray was likewise chosen to be the chief designer, but he was still in the process of completing an early version of the 1103-based Naval Tactical Data System (NTDS), and he did not leave Sperry to join CDC until it was complete.

Early designs and Cray's big plan

CDC started business by selling subsystems, mostly drum memory systems, to other companies. Cray joined the next year, and he immediately built a small transistor-based 6-bit machine known as the "CDC Little Character" to test his ideas on large-system design and transistor-based machines. "Little Character" was a success, and CDC soon released a 48-bit transistorized version of their 1103 re-design as the CDC 1604 in 1959, with the first machine delivered to the U.S. Navy in 1960. Legend has it that the 1604 designation was chosen by adding CDC's first street address (501 Park Avenue) to Cray's former project, the ERA-Univac 1103. A 12-bit cut-down version was also released as the CDC 160A in 1960, arguably the first minicomputer. The 160A was particularly notable as it was built as a standard office desk item, which was a rather-unusual packaging for that era. New versions of the basic 1604 architecture were rebuilt into the CDC 3000 series, which sold through the early and mid-1960s.

Cray immediately turned to the design of a machine that would be the fastest (or in the terminology of the day, largest) machine in the world, setting the goal at 50 times the speed of the 1604. This required radical changes in design, and as the project "dragged on" — it had gone on for about four years by then — the management got increasingly upset and it demanded greater oversight. Cray in turn demanded (in 1962) to have his own remote lab, saying that otherwise, he would quit. Norris agreed, and Cray and his team moved to Cray's home town, Chippewa Falls, Wisconsin. Not even Bill Norris, the founder and president of CDC, could visit Cray's laboratory without an invitation. (See story of a salesman's uninvited visit to Chippewa Falls here .)

Peripherals business

Through the 1960s, Norris became increasingly worried that CDC had to develop a "critical mass" in order to compete with IBM. In order to do this, he started an aggressive program of buying up various companies to round out CDC's peripheral lineup. In general, they tried to offer a product to compete with any of IBM's, but running 10% faster and costing 10% less. This was not always easy to achieve.

One of its first peripherals was a tape transport, which led to some internal wrangling as the Peripherals Equipment Division attempted to find a reasonable way to charge other divisions of the company for supplying the devices. If the division simply "gave" them away at cost as part of a system purchase, they would never have a real budget of their own. Instead, a plan was established in which it would share profits with the divisions selling its peripherals, a plan eventually used throughout the company.

The tape transport was followed by the 405 Card Reader and the 415 Card Punch , followed by a series of tape drives and drum printers, all of which were designed in-house. The printer business was initially supported by Holley Carburetor in the Rochester, Michigan suburb outside of Detroit. They later formalized this by creating a jointly-held company, Holley Computer Products . Holley later sold its stake back to CDC, the remainder becoming the Rochester Division.

Norris was particularly interested in breaking out of the punched card–based workflow, where IBM held a stranglehold. He eventually decided to buy Rabinow Engineering, one of the pioneers of optical character recognition (OCR) systems. The idea was to bypass the entire punched card stage by having the operators simply type onto normal paper pages with a "known" typewriter font, and then submit those pages to the computer. Since a typewritten page contains much more information than a punched card (which has essentially one line of text from a page), this would offer savings all around. Unfortunately, this seemingly-simple task turned out to be much harder than anyone expected, and while CDC became a major player in the early days of OCR systems, it has remained a niche product to this day. Rabinow's plant in Rockville, MD was closed in 1976, and CDC left the business.

With the continued delays on the OCR project, it became clear that punched cards were not going to go away any time soon, and CDC had to address this as quickly as possible. Although the 405 remained in production, it was an expensive machine to build. So another purchase was made, Bridge Engineering, which offered a line of lower-cost as well as higher-speed card punches. All card-handling products were moved to what became the Valley Forge Division after Bridge moved to a new factory, with the tape transports to follow. Later on, the Valley Forge and Rochester divisions were spun off to form a new joint company with National Cash Register (later NCR Corporation), Computer Peripherals Inc (CPI), in order to share development and production costs across the two companies. ICL later joined the effort. Eventually the Rochester Division was sold to Centronics in 1982.

Another side-effect of Norris's attempts to diversify was the creation of a number of service bureaus that ran jobs on behalf of smaller companies that could not afford to buy computers. This was never very profitable, and in 1965, several managers suggested that the unprofitable centers be closed in a cost-cutting measure. Nevertheless, Norris was so convinced of the idea that he refused to accept this, and ordered an across-the-board "belt tightening" instead.

CDC 6600: defining supercomputing

Meanwhile at the new Chippewa Falls lab, Seymour Cray, Jim Thornton, and Dean Roush put together a team of 34 engineers, which continued work on the new computer design. In 1964, this was released onto the market as the CDC 6600, out-performing everything on the market by roughly ten times. The 6600 had a CPU (Central Processing Unit) with multiple, asynchronous functional units, and it used 10 logical, external I/O processors to off-load many common tasks. That way the CPU could devote all of its time and circuitry to processing actual data, while the other controllers dealt with the mundane tasks like punching cards and running disk drives. Using late-model compilers, the machine attained a standard mathematical operations rate of 500 kilo-FLO