Assessment of fault-tolerant computing systems at NASA´s Langley Research Center

In the early 1970´s while NASA was studying Advanced Technology Transport concepts, researchers at NASA´s Langley Research Center (LaRC) recognized that digital computer systems would be controlling civil transport aircraft in the near future and that the technology did not exist to determine if these digital systems would be reliable enough for this role. In addition, although several existing computer system concepts showed promise to meet the civil transport requirements, none had been realized in an operational system. A multi-initiative program was developed to determine how to assess reliability and performance of fault-tolerant digital computer systems for determining if they could meet the requirements of a civil transport. Subsequent research emphasized the application of formal methods, system safety and digital upset. Some results indicated that dissimilar software may not be reliable enough for critical applications, testing alone will not prove the reliability of highly reliable digital systems and formal methods can find design errors missed by other assessment techniques. Future research will center around the application of formal mathematical methods, insuring software safety, and determination of digital system upsets due to electromagnetic radiation. The long term goal is to define methods for producing error-free systems for flight crucial civil transport applications