IBM and Apollo 11: Putting man on the moon

 

IBM and Apollo 11: The partnership that led to the first man on the moon

It’s been fifty years since Apollo 11 landed the first man on the moon.

To achieve this, some four thousand IBM technicians were involved in the Apollo programme.

Pioneering the technologies, building the computers, and writing the software programmes that first launched the missions, then later guided its astronauts safely back to Earth.

From networking computers to processing unprecedented amounts of data, NASA and IBM were the first to face challenges now common in our modern world.

 

IBM and Apollo 11 – The beginnings

The 27 kg guidance computer launched into space on a Gemini mission of 1965 demonstrated that advanced technologies would be needed to get astronauts to the moon and back.

The astronauts, all previously aircraft pilots, wanted to be in control of the mission from start to finish, including the launch.

But NASA knew that things happen so fast in a launch that human reactions just can’t keep up.

What’s more, a lunar mission would involve two spacecraft flying together a quarter-million miles away.

There was going to be a whole lot of data that needed processing.

And it would need more than just computing power on the spacecraft.

It would need a sophisticated and powerful network.

One that could manage, work with, and transfer all the information required to the spacecraft, as well as receive everything beamed back to Earth.

 

IBM and Apollo 11 – New systems

IBM’s 7090 series of mainframe computers formed the backbone of NASA’s mission control and data management for the Mercury and Gemini programmes.

Then in 1964 IBM unveiled its latest series of computers, System 360.

At one million bytes, the memory was four times greater than that of the Gemini-era computers. The new OS/360 operating system could also multitask.

 

These computers weren’t just singular systems: they could be networked together to build a multiprocessing system. A compatible family of machines, pioneering the eight-bit byte and supporting different processing speeds, they could talk to and work with each other.

More than just automating a task, a whole system of computers could manage complex processes.

Even so, the Apollo 11 mission was expected to generate more data than the computers could handle.

To eke out as much memory space as they could, IBM engineers developed a shorthand code rather than using full sentences, shaving off words wherever possible.

 

Even in shorthand, however, it would require a circuit capable of transmitting a novel a minute.

The IBM computing and data processing system was set up at the Real-Time Computer Complex (RTCC) in Houston, Texas.

It would be expected to collect, process and send to Mission Control all the information needed to direct every phase of the Apollo 11 mission.

 

IBM and Apollo 11 – Facing challenges

Under the enormous pressure of ensuring everything was perfect, the IBM technicians had already worked every day for months, sleeping little as they sought to overcome challenges never faced before.

They were writing code, programming computers, and running simulations for the computers and software required at multiple locations, including Cape Canaveral in Florida and Mission Control Center in Houston, Texas.

But on July 16 1969, the day of the launch, they would all find themselves working harder than ever.

On this day, and throughout the flight of Apollo 11, IBM and NASA personnel tirelessly worked side-by-side.

IBM technicians sat alongside mission controllers at NASA’s Manned Spacecraft Center (now the Lyndon B. Johnson Space Center) in Houston, reading data and making minute-by-minute analyses.

IBM engineers were also on hand at the Goddard Space Flight Center in Maryland, home of the worldwide network of relay and tracking stations.

Were they aware that the systems, software and operations they’d developed would lay the groundwork for mission-critical systems widely used in business today?

 

IBM and Apollo 11 – The Apollo 11 rocket’s brain

IBM engineers developed miniaturised circuity reducing a refrigerator-sized mainframe into a suitcase-sized computer.

It had to be small enough to fit inside the Saturn V, the Apollo 11 rocket.

Built at the Marshall Space Flight Center in Huntsville, Alabama, the Saturn Instrument Unit (IU) was shipped to the Kennedy Space Center, Florida, for final testing and integration into the rocket.

The nerve centre for the launch vehicle, the IU could precisely sense altitude, acceleration, velocity and position.

It determined when to fire the three rockets, when to jettison them, and where to point the spacecraft, setting the desired course with control signals.

Data was being constantly sent to five System 360 machines on the ground.

These absorbed, calculated, evaluated, and translated all this information into meaningful information the NASA and IBM technicians used to monitor the rocket’s trajectory.

 

Error, Error!

The Apollo Guidance Computer had only one error message.

When it flashed, it meant the end really was near.

Even so, 85% of the code written for Apollo 11 was there to anticipate the “unknown unknowns”.

This was all quite literally uncharted territory. And the IBM engineers were determined to be prepared for any eventuality.

As the Eagle landing craft was preparing to land on the moon, an error message did flash.

The Apollo 11 spacecraft crew had mistakenly left a radar system on while the landing crew used their own, calculating the distance to the ground.

The computer was taking in too much data to function.

Fortunately, the operating system in the Apollo 11 spacecraft could multi-task eight jobs at a time.

The Apollo system also implemented a virtual machine capable of performing advanced mathematics and offering complex instructions.

And on July 20, 1969, the world watched as Neil Armstrong safely brought the landing craft down to be the first man on the moon.

The NASA and IBM technicians breathed a sigh of relief, celebrated – and in some cases slept for the first time in days.

 

IBM and Apollo 11 – Moon landing facts

When did Apollo 11 launch? On July 16 1969

When did Apollo 11 land on the moon? On July 20 1969

What was the top speed of Apollo 11? 7 miles per second (to break free of the Earth’s gravitational field)

How much would the moon landing cost today? Around $150 billion ($25.4 billion in 1969)

What kind of computing power was utilised? Apollo’s Guidance Computer had 32k of storage, 2k of memory, and ran at 1.024 MHz

 

IBM and Apollo 11 – and beyond

The System 360 computers processed the data for the first lunar landing from 240,000 miles away in Houston.

It also calculated the lift-off data required for astronauts Neil Armstrong and Edwin “Buzz” Aldrin to rendezvous back with the command module piloted by Michael Collins.

It then plotted the flight back to Earth.

Despite this remarkable achievement, perhaps System 360’s true legacy is the beginning of computer compatibility.

IBM recently helped launch the first AI-powered astronaut assistant, CIMON.

When astronaut’s in the International Space Station ask CIMON a question, its IBM Watson capabilities ensure it understands both context and the intention behind it.

Where will AI technologies take us next?

To find out, explore IBM Watson Assistant here.

Related content

Resource Centre

Transform your business with next-level analytics

Free Trial / Demo

Empowering machine learning and AI to catalogue and analyse data

Blog

How to Scale the AI ladder

Back to top