Find the answer to your Linux question:
Results 1 to 2 of 2
Enjoy an ad free experience by logging in. Not a member yet? Register.
  1. #1

    [SOLVED] Is Fault-Tolerance subject wear out?


    1 - I was reading some theory stuff about fault-tolerance techniques and the ultimate fault-tolerance technique is making a system byzantine fault-tolerance. Since some guy already proved how to make a system byzantine fault-tolerant, this means that doesn't exist nothing more to do in this area? Is everything discovered about fault-tolerance?

    2 - I've the idea that doesn't exist any corporation who cares about fault-tolerant systems. Am I right? Is this area neglected or totally discovered?


  2. #2
    Linux Guru Rubberman's Avatar
    Join Date
    Apr 2009
    I can be found either 40 miles west of Chicago, in Chicago, or in a galaxy far, far away.
    1. There is always room for research into these topics. Just because it has (possibly) been proven that one can build a byzantine fault-tolerant system, doesn't mean it can be done in practical terms. In any case, solutions to this are similar in structure to the distributed transaction two-phase commit algorithms. Anyway, I'm sure you can find a PhD thesis subject in this area without problem!

    2. Wrong. A lot of major organizations are very concerned about this subject, and they spend a lot of money on fault-tolerant hardware, software, networks, etc. True fault-tolerant systems are very expensive, and generally are not as fast as other systems. I have spent a good portion of my professional career designing, developing, and proving what I call fault-resilient systems. Failure states are tolerated, but recoverable with fail-over hardware and software. No single point of failure. Even these are difficult to design and build, but a lot of manufacturers of high-value products, such as semiconductor makers, require 24x365 operation of their factories - down time runs millions of dollars per hour in losses (about $10M for a 300mm fab). As the saying goes, stuff happens. So, we designed these systems to tolerate a fair amount of hardware and/or software failures yet continue to run efficiently enough that the production line was not impacted in any significant way.
    Sometimes, real fast is almost as good as real time.
    Just remember, Semper Gumbi - always be flexible!

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts