Atlas (computer)

The Atlas was a giant computer.

It was very fast. It was one of the best in the world.

Many people worked to build it. They used small parts called transistors. These parts helped the machine run.

The computer had a special trick. It used a new way to store data. This helped it do many jobs at once.

It was so big and strong. Some said it held half the computer power in its land. It was a very important machine.

Scientists still study how it worked. It helped change how all computers work today.

The Atlas was one of the first supercomputers. It was used from 1962 to 1972. People said it was very powerful. If it stopped working, half of the computer power in the United Kingdom was lost!

Many people worked to make it. The University of Manchester, Ferranti, and Plessey joined together. They used small parts called transistors. These parts helped the machine run fast.

Atlas had a very special way to work. It used virtual memory. This is a way to use more memory than the machine actually has. It used a method called paging to do this. This idea was so good that almost all computers use it today!

Other machines were built from the Atlas design. One was called Titan. Another was called Atlas 2. These machines helped different groups do science work. Even today, we remember Atlas. It helped change how all computers work.

The Atlas was one of the world's first supercomputers. It was a very powerful machine used from 1962 to 1972. It was so important that people had a special saying about it. They said if Atlas went offline, half of the United Kingdom's computer capacity was lost!

This machine helped change how we use technology today. It was a second-generation computer that used small parts called germanium transistors. These parts helped the computer work much faster than older models.

Atlas worked in a very clever way using virtual memory. At that time, people called this a "one-level store." It used a method called paging to manage information. Think of paging like a library that uses small cards to find books. This allowed the computer to use more memory than it actually had inside. The system used a special program called a Supervisor to coordinate everything. This Supervisor was a huge job to build. It used 35,000 lines of code to help the machine handle many tasks at once.

Many smart people worked together to build this giant machine. The University of Manchester, Ferranti, and Plessey joined forces for the project. A team led by Tom Kilburn started experimenting with transistors in 1956. They originally wanted to build a machine called MUSE, which stood for "microsecond engine." They wanted it to be fast enough to do one task every microsecond. Later, the project was renamed Atlas. A man named David Howarth led a large team of six programmers to finish the complex software.

There were several different Atlas machines built for different places. The first one was officially started on December 7, 1962. Another large version was built at the Atlas Computer Laboratory near Oxford. This one had 32 tape drives and a lot of memory. Ferranti also made a version for the University of Cambridge called Titan. This machine was also known as Atlas 2. Other Atlas 2 machines went to the CAD Centre and the Atomic Weapons Research Establishment.

Even though Atlas is old, its ideas are still used in your gadgets. The way it used virtual memory is now used in almost every computer. Some parts of the Atlas are even kept in museums today. National Museums Scotland keeps parts of the Chilton Atlas in Edinburgh. The main control console was found again in July 2014. It is now located at the Rutherford Appleton Laboratory near Oxford. We can still look at these parts to see how computing began.

The Atlas was one of the world's first supercomputers. It operated from 1962 until 1972. This machine was so massive in its capability that a famous saying emerged. People claimed that if Atlas went offline, half of the United Kingdom's total computer capacity was lost.

As a second-generation computer, it utilized discrete germanium transistors. These components allowed it to perform much faster than previous generations of technology. The Atlas was not just a machine; it was a landmark in computing history.

The most significant breakthrough of the Atlas was its use of virtual memory. At the time, engineers called this concept a "one-level store." It used a specific technique known as paging to manage data. Paging allowed the computer to use more memory than was physically present in its core store. The system integrated 16K words of core store with 96K words of drum store. This was achieved by treating the drum store as an extension of the main memory. This approach was revolutionary and eventually became a standard feature in almost all modern computers.

To manage these complex tasks, the Atlas relied on a sophisticated operating system called the Supervisor. Building this software required a massive effort by a team of programmers. Led by David Howarth, the team produced 35,000 lines of assembler language. This software enabled multiprogramming, which allowed the machine to handle many different tasks at once. This was essential because the computer needed to coordinate various peripheral devices. Without the Supervisor, the central processing unit (CPU) would waste time waiting for data from slow devices like card readers.

The development of Atlas began as a response to American computing progress. In the mid-1950s, the United Kingdom realized it was falling behind the United States. Tom Kilburn's team at the University of Manchester began experimenting with transistor-based systems in 1956. They initially named their design MUSE, which stood for "microsecond engine." This name reflected a goal to perform one instruction every microsecond. Eventually, the project became a joint effort between the University of Manchester, Ferranti, and Plessey. Ferranti received a grant of £300,000 from the National Research Development Corporation to build the system.

Several different versions of the Atlas were installed in various locations. The first machine was officially commissioned on December 7, 1962. A second, very large installation was built at the Atlas Computer Laboratory near Oxford. This version contained 48 kWords of 48-bit core memory and 32 tape drives. Ferranti also developed a derivative system for the University of Cambridge. This machine was called Titan, or Atlas 2, and featured a different memory organization. Other Atlas 2 machines were delivered to the CAD Centre and the Atomic Weapons Research Establishment.

In terms of raw power, Atlas was a true leader of its era. While American machines like IBM STRETCH were slightly faster, Atlas was much faster than the Univac LARC. For example, Atlas could add two floating-point numbers in about 1.59 microseconds. The STRETCH machine completed the same task in 1.38 to 1.5 microseconds. Despite this, experts claimed Atlas was the most powerful system because of how it handled large sets of jobs. By the mid-1960s, the original Manchester machine was running on a schedule of 20 hours per day. During this time, it might run as many as 1,000 different programs.

The legacy of the Atlas continues to influence the field of computer science. Its invention of virtual memory is recognized as a major milestone by organizations like the IEEE. Even the design of later machines was influenced by the Atlas. The CDC 6600, which eventually surpassed the Atlas, was developed using ideas from the original MUSE design. Today, pieces of this history are preserved for future generations. National Museums Scotland holds parts of the Chilton Atlas in Edinburgh. The main control console was rediscovered in 2014 and is now at the Rutherford Appleton Laboratory.