Project Gemini

Through its work on the Vanguard satellites and Project Mercury, IBM had gained a reputation within NASA for providing hardworking personnel and top-of-the-line data processing machines. So when it came time to select a contractor for the computer systems for Gemini, IBM was already a favorite. After submitting a proposal to the agency’s open call for bids, the company was officially awarded the contract. And with that, IBM set to work.

In many ways, NASA designed Gemini to test the agency’s readiness for the Apollo missions. The team needed to ensure that the spacecraft could sustain flight for a trip to the moon and back. This required monitoring any possible long-term effects of extended spaceflight on the astronauts’ health and mental well-being. Moreover, the team needed to be able to maneuver the spacecraft into a rendezvous — coming within a close distance of a launch vehicle or sister craft in the same orbit — and to fine-tune re-entry and landing methods. IBM’s role was to provide the computing power to achieve these milestones.

Originally, the IBM team proposed a system comprising three 7094 computers. Together they would run a mission control program and a simulation computer program. It quickly became clear, however, that this schema didn’t have the data processing prowess that would be needed for the more complex Gemini and Apollo missions.

The team drew up a five-computer system, which would come to be known as the Real-Time Computer Complex. Within this architecture, the first four computers were designated as: the Mission Operational Computer, the Dynamic Standby Computer, the Simulation Operations Computer and the Ground System Simulation Computer. The last was a standby for future software development. During the missions, these machines would perform a staggering 25 billion calculations every 24 hours, providing flight controllers with nearly instantaneous reports on the crafts’ progress.

In addition to the ground-level system, IBM developed an onboard guidance computer. Capable of storing nearly 20,000 bytes of information and performing 7,000 calculations a second, it enabled astronauts to perform their own calculations during flight for the first time. Roughly the size of two mailboxes side by side, and weighing just under 60 pounds, the machine was engineered to process data from onboard and ground control systems — as well as from the astronauts themselves — in order to deliver precise navigational and support information to the astronauts on board. This gave the pilots the power to guide and maneuver their spacecraft on their own.

This onboard guidance computer would serve as the brains behind the first ever spacecraft rendezvous, when Gemini 6 and Gemini 7 met nose to nose, 120 feet apart, 185 miles above Hawaii. It also processed the complex calculations that brought Gemini 11, on its way back to Earth, closer to the recovery carrier than any previous mission.

The software developed for Gemini eventually gave rise to one of the largest computer programs in history. And it continued to evolve throughout the mission’s duration. For example, in the latter stages, all data processed by the tracking network became digital, which greatly increased the software’s productivity. The IBM 2361 core storage unit, meanwhile, developed near the end of the program, offered up a storage capacity of 2.6 megabytes of data, making it the largest computer memory built by the company up until that point.

Pushed to new limits by rendezvous and orbit-change operations, IBM’s spaceflight software grew increasingly sophisticated and effective over the course of the Gemini campaign and laid the groundwork for the even more ambitious ventures to come.

By the time Gemini came to a close in 1966, the NASA and IBM teams had achieved several historic firsts:

• The first maneuvers by an orbiting spacecraft
• The first rendezvous in space
• The first docking in space
• The first navigation in space
• The first rendezvous in initial orbit
• The first onboard computer-controlled reentry

Perhaps most importantly, the teams had achieved their true goal — to make it obvious that a lunar mission was well within humanity’s grasp.