CDC 6000 series

The CDC 6000 series was a family of very fast and powerful computers made by Control Data Corporation in the 1960s. These computers included models like the CDC 6200, CDC 6300, CDC 6400, CDC 6500, CDC 6600, and CDC 6700. They were huge machines that could handle many different jobs at once, making them very useful for both scientific research and business tasks.

Each computer in the CDC 6000 series was built with solid-state technology, meaning they didn’t use old vacuum tubes. They could run many programs together thanks to a special system called multiprogramming and could even work on several tasks at the same time using multiprocessing. They also allowed people to share the computer’s power through time-sharing and manage data efficiently with data management systems.

These computers were some of the first ever supercomputers, and the CDC 6600 was the most famous of them all. They used an operating system called SCOPE, short for Supervisory Control Of Program Execution, to keep everything running smoothly. By 1970, there was also another system called KRONOS designed for sharing the computer’s resources among many users. The CDC 6000 series helped change the way people did big calculations and handled large amounts of information.

Overview

The CDC 6000 series computers are made up of four main parts: central memory, one or two fast central processors, ten peripheral processors (Peripheral Processing Unit), and a display console. These computers have a design where tasks are spread out across different parts.

The different models in the series mainly vary by the number and type of central processors they have. The CDC 6400 has one central processor that can only handle one instruction at a time. The CDC 6500 has two central processors, both like the CDC 6400. The CDC 6600 has one central processor with ten parts that can work on instructions all at once. The CDC 6700 also has two central processors, one like the CDC 6600 and one like the CDC 6400.

Some features from an earlier series of computers, the CDC 3000, were also used in the CDC 6000 series. For example, the way it does math and the names of its programming languages and operating systems.

Today, the only CDC 6000 series computer still running is a CDC 6500. It was built in 1967 and used by Purdue University until 1989. It was then given to a museum before being sold at an auction in 2024.

History

The first computer in the CDC 6000 series was the supercomputer CDC 6600, created by Seymour Cray and James E. Thornton in Chippewa Falls, Wisconsin. It was introduced in September 1964 and could perform up to three million instructions per second, making it three times faster than the IBM Stretch. It stayed the fastest machine for five years until the CDC 7600 came out. This computer used Freon to stay cool.

Control Data made around 100 of these machines, each costing between $6 to $10 million. Later, they introduced the CDC 6400 in April 1966, which was slower and less expensive. It followed a design with serial processing instead of parallel processing. After that, they made the CDC 6500 in October 1967, which had two 6400-style processors. Finally, the CDC 6700 was released in October 1969, combining both 6600-style and 6400-style processors.

Hardware

Central memory (CM)

The CDC 6000 series computers could run up to seven programs at the same time in central memory. These programs sent their instructions to the central processor, which then carried out the tasks and sent the results back.

Information was stored in central memory as "words," each containing 60 binary digits, or bits. Moving this information in and out of memory was very fast, taking only 100 nanoseconds.

Extended Core Storage (ECS)

ECS added more memory space to the CDC 6000 series, boosting their computing power. It used special storage banks with buffers to improve speed in some tasks, though regular memory was often faster.

Central processor

The central processor was the main engine of the computer, handling calculations and tasks like addition, subtraction, and logical operations. It did not manage input or output; that was done by other parts of the system called peripheral processors.

The processor had 24 registers—special storage spots—for handling data. Instructions could be 15 or 30 bits long, and the processor could handle multiple tasks efficiently. Different models in the series had varying abilities, with some able to work on several tasks at once.

Peripheral processors

Up to ten peripheral processors shared the central memory with the main processor. These smaller computers had their own memory and handled tasks like reading from disks or magnetic tapes, freeing the main processor to focus on calculations.

Data channels

Peripheral processors communicated with devices like disks and tapes through special links called data channels. Only one processor could use a channel at a time, but they could pass data between each other.

Display console

Operators controlled the system using a console with two screens, a big step ahead from older systems with many lights and switches. The console showed what was happening in the computer and let operators give commands. It even had early games like a baseball game and a Pac-Man-like game.

Minimum configuration

A basic CDC 6000 system needed central memory, storage devices like disks or tapes, a card reader, a printer, and tape units. Bigger systems could include more memory, extra storage, and other devices.

Peripherals

The CDC 6000 series could connect to many devices, including card readers, disk systems, magnetic tape units, communications multiplexers for remote jobs and timesharing, and data channel converters.

Versions

"CDC 6400" redirects here. Likewise: CDC 6500, and CDC 6700."CDC 6600" has its own article.

The CDC 6600 was the most advanced model. The CDC 6400 was slower and cheaper.

The CDC 6500 had two CPUs like the 6400, meant for tough problems. The CDC 6700 also had two CPUs, one from the 6600 and one from the 6400. The CDC 6415 was a cheaper, slower model with fewer parts. The CDC 6416 was an upgrade for better performance.

The 6600

Main article: CDC 6600

The CDC 6600 is the main mainframe supercomputer in the 6000 series made by Control Data Corporation. It was thought to be the first successful supercomputer, working three times faster than its predecessor, the IBM 7030 Stretch. It could handle up to three megaFLOPS, and about 100 were sold. It was the fastest computer in the world from 1964 to 1969.

The CDC 6600 was one of the first computers to use a special design idea and a unique way to handle numbers. It was also the first common computer to have a specific setup for moving data.

The first CDC 6600s were delivered in 1965 to labs in the United States. One went to a university in New York City. The first outside the US went to a lab near Geneva, Switzerland, where it helped study photos from experiments. In 1966 another went to a university in California, and another to a university in Texas.

A CDC 6600 is displayed at the Computer History Museum in Mountain View, California.

The 6400

The CDC 6400, part of the CDC 6000 series, is a mainframe computer made by Control Data Corporation in the 1960s. Its main part worked similarly to the CDC 6600. Unlike the 6600, which could handle many instructions at once, the 6400 could only handle one at a time. This made it slower but also cheaper. Other parts like memory and input/output were the same as the 6600.

In December 1966, a CDC 6400 started operating at a university in California.

In 1966, a university in Germany got a CDC 6400, the first of its kind in Germany and the second in Europe after a lab in Switzerland.

Dual-CPU systems

The 6500

The CDC 6500, with a dual-CPU 6400, is the third supercomputer in the 6000 series made by Control Data Corporation and designed by supercomputer expert Seymour Cray. It was announced in 1964 and the first one was delivered in 1967.

It had twelve separate computers. Ten were for helping tasks, each with its own memory and could run programs on their own, along with two main 6400 processors. Instead of air cooling, it used liquid cooling, with three separate cooling units.

CDC 6500 systems were installed at:

• Purdue University
• Michigan State University
• CERN
• the technical lab at the Patrick Air Force Base in 1978
• the Laboratory of Computing Techniques and Automation in the Joint Institute for Nuclear Research (USSR)
• University of Colorado Boulder