The Space Shuttle program
IBM computers and software powered NASA’s Space Shuttle program through more than a hundred successful flights
The Space Shuttle blasting off

After the rapid aerospace advancements of the 1950s and ’60s, the Space Shuttle program marked the start of a new, more sustainable phase of space exploration. Launched in 1972, NASA’s fifth human spaceflight initiative envisioned a cost-effective regimen of otherworldly scientific experimentation centered on a vehicle conceived to make the journey into space as many as a hundred times.

The Space Shuttle comprised three main elements: the airplane-like craft, known as the orbiter; a larger external fuel tank filled with liquid propellant; and two solid rocket boosters. Only the orbiter and the rocket boosters were reusable. Over the course of five decades, five generations of the vehicle — each weighing more than 2,200 tons — would routinely transport satellites and military hardware, as well as astronauts and scientists, into space and back.

Every mission was supported by IBM computers, software and employees.

Shuttle support
Next-gen computers for a next-gen spacecraft

Previous to the Space Shuttle program, IBM had contributed technical expertise and tools to each of NASA’s manned missions — Mercury, Gemini and Apollo — as well as to early US government satellite projects in the late 1950s. For this particular contract, the agency tasked the company with providing an expansive system to cover everything from flight control to monitoring and navigation — and enlisted the efforts of IBM teams across the country.

The orbiter presented significant challenges from the outset. To troubleshoot the first-of-its-kind vehicle, the NASA team designed an experimental craft, named Enterprise, in 1974 to trial the new concept via a series of ground and aerial test flights within Earth’s atmosphere. Construction on Columbia, the agency’s first operational Space Shuttle, began less than a year later.

The Space Shuttle’s computers, displays and software programs — both for onboard and at mission control — were significant departures from the systems employed in previous generations. While earlier crafts depended almost exclusively on a few large-scale digital computers, the Space Shuttle incorporated more than 150 mini computers as well as 300 microprocessors and video displays — in eight colors, rather than the customary black and white.

The reliance on mainframes, however, remained. In Owego, New York, at the Federal Systems Division (FSD), IBMers devised a complex, onboard data processing system around a five-computer circuit. Built on System/360 mainframes and weighing 55 pounds each, these general purpose computers, or GPCs, belonged to IBM’s Advanced System/4 Pi (AP-101) avionics computer series. During missions, they were capable of handling flight control, communications and navigation while processing half a million operations a second. Meanwhile, IBM custom built input/output processors to facilitate communication between onboard computers and the orbiter’s other systems.

Every mission was supported by IBM computers, software and employees
A fail-safe system
Cross-checks, redundancy and fly-by-wire

The Space Shuttle’s GPCs performed 500 cross-checks every second to ensure redundancy and prevent failure. If a probe exposed differing information, the at-fault machine would be automatically replaced by a spare computer, the system’s fifth unit.

Through a concept known as fly-by-wire, the computers controlled the orbiter during flight via electric signals sent directly to hydraulic-driven actuators. During noncritical periods in orbit, a single GPC operated the Space Shuttle without assistance. During critical phases, four primary computers performed 325,000 operations per second to execute flight navigation. Meanwhile, at the Shuttle Data Processing Complex in Texas, three IBM System/370 Model 168 computers provided flight support for ground control.

Also in Texas, IBMers were busy devising the software for these onboard units. The most sophisticated space vehicle programs developed up to that point covered everything from flight controls and crew displays to monitoring, guidance and navigation. For the launch facilities and preparation in Florida, IBMers bolstered these programs with the so-called Launch Processing System — a complex test data arrangement that integrated a minicomputer network and 10,000 lines of microcoded instructions for each Shuttle Firing Room. The company also supplied NASA with the displays for the laboratory, simulator, spare systems and the Space Shuttle’s onboard communication.

When the time came for Columbia to embark on its maiden voyage in April of 1981, IBM hardware and software were there to manage and monitor each stage, from liftoff to landing.

Powering the Space Shuttle
From mainframes to Thinkpads

During the 1980s and ’90s, the system developed by IBM for NASA powered dozens of flights, encompassing the Challenger, Discovery, Atlantis and, finally, the Endeavour missions. Throughout these phases, the company updated and optimized the Space Shuttle’s computers and programs as technology advanced.

In 1991, the newly developed IBM AP-101 onboard flight computers replaced the original AP-101B, more than doubling the memory availability of the GPC and boosting CPU speed. In 1993, during an assignment to refurbish the Hubble Space Telescope, Endeavour launched with an IBM ThinkPad 750C onboard — the first notebook computer to enter space. Astronauts used it during the mission to analyze color images and sketches of the telescope that were loaded onto the computer’s hard drive.

In the years that followed, several iterations of the compact ThinkPad filled ever more critical roles both in Space Shuttle missions and, later, for work aboard the International Space Station.

Tragedy and rebirth
A space travel revolution comes to a close

While the Space Shuttle program was largely considered a great success over the course of 133 completed flights, it was of course not without failure. The first tragedy came in 1986, with the fatal, midair destruction of the Challenger — caused by structural failure in one of the solid rocket boosters, taking the lives of five NASA astronauts and two payload specialists, including a schoolteacher, Christa McAuliffe. Then in 2003, following a 16-day scientific mission, the Columbia orbiter broke apart while re-entering Earth’s atmosphere. It was the result of damage to the wing caused by a piece of foam, dislodged during liftoff. All seven crew members lost their lives. The deceased have been memorialized in several monuments across the US, including in brass-plated gravestones at Arlington National Cemetery. After the Columbia disaster, NASA spent several years running its orbiters through vigorous safety checks. The missions resumed in 2005 and continued for six more years, until being discontinued in 2011.

Through its revolutionary vision for near-space travel, the Space Shuttle program heralded both an adventurous new era of aerospace exploration — and the next, laureled chapter of IBM and NASA’s decades-long partnership. “We started in the Mercury program, had the first onboard computers in the Gemini program, provided the IU for the Apollo program,” FSD Houston general manager Jack Winter explained in 1985. “We’ve been in Mission Control Center since day one, provided computers and software for Skylab. We were a part of the Apollo-Soyuz program, and now, of course, Shuttle.”

“It’s provided a great source of pride for the IBM Corporation,” he continued,“as well as being a service to the national interest.”

Related stories Project Mercury

When it came to sending an American into space for the first time, NASA relied on IBM technology and personnel

Project Gemini

IBM developed a system to support NASA’s program for a second crewed spaceflight and make the case for sending humans to the moon

Apollo

A diverse group of 4,000 IBMers helped NASA achieve one of the greatest feats of human history — sending astronauts to the moon