14 DAY TRIAL // The new computer which Google thinks will spark a new innovation is "mathematically" guaranteed not to lose information if a crash occurs, or by accident.
Conceived by MIT's Computer Science and Artificial Intelligence Lab (CSAIL), the new system creates the foundation for a more advanced and faster design since the prototype is much slower than the more powerful but less failsafe current computers.
CSAIL researchers worked hard to make sure the system is reliable both when they're operating normally, but also when the system crashes, has a power failure or is plagued by bugs or hardware errors. The current issue of recovering data from crashes is that too many times crashes have affected the exact areas that data needs to be recovered from, and although some places might seem secure now, they might not be next time, even in very well tested file systems.
Safer computers for everybody
The team tested the computer's abilities through a process known as formal verification, something that describes mathematically the operational limits for a computer program, something that is meant to prove the program cannot exceed those limits. This way the computer is not crash safe, but it can save data in the process.
For Ulfar Erlingsson, the lead manager for security research at Google, the bright side of this research is that everything developed and researched at MIT can be easily duplicated in test cases, without being so "esoteric" in nature that it remains something tightly linked to the MIT environment. Apparently, the new design is user-friendly enough to start a wider innovation among business computer manufacturers that will jump on the occasion to safeguard their computers against such system crashes.
The guys at MIT hope the new system safeguard design will be applied in many different domains in the IT and business-to-business industries. The new technology will be unveiled at a symposium later this year.