Triple modular redundancy
Triple modular redundancy, sometimes called triple-mode redundancy, (TMR) is a fault-tolerant form of N-modular redundancy, in which three systems perform a process and that result is processed by a majority-voting system to produce a single output. If any one of the three systems fails, the other two systems can correct and mask the fault.
Some ECC memory uses triple modular redundancy hardware (rather than the more common Hamming code), because triple modular redundancy hardware is faster than Hamming error correction software. Space satellite systems often use TMR, although satellite RAM usually uses Hamming error correction.
Some communication systems use N-modular redundancy as a simple form of forward error correction. For example, 5-modular redundancy communication systems (such as FlexRay) use the majority of 5 samples – if any 2 of the 5 results are erroneous, the other 3 results can correct and mask the fault.
Modular redundancy is a basic concept, dating to antiquity, while the first use of TMR in a computer was the Czechoslovak computer SAPO, in the 1950s.
To use triple modular redundancy, a ship must have at least three chronometers; two chronometers provided dual modular redundancy, allowing a backup if one should cease to work, but not allowing any error correction if the two displayed a different time, since in case of contradiction between the two chronometers, it would be impossible to know which one was wrong (the error detection obtained would be the same of having only one chronometer and checking it periodically). Three chronometers provided triple modular redundancy, allowing error correction if one of the three was wrong, so the pilot would take the average of the two with closer reading (vote for average precision).
There is an old adage to this effect, stating: "Never go to sea with two chronometers; take one or three."
Mainly this means that if two chronometers contradict, how do you know which one is correct? At one time this observation or rule was an expensive one as the cost of three sufficiently accurate chronometers was more than the cost of many types of smaller merchant vessels. Some vessels carried more than three chronometers – for example, the HMS Beagle carried 22 chronometers. However, such a large number was usually only carried on ships undertaking survey work as was the case with the Beagle.
Majority Logic Gate
In TMR, three identical logic circuits (logic gates) are used to compute the same set of specified Boolean function. If there are no circuit failures, the outputs of the three circuits are identical. But due to circuit failures, the outputs of the three circuits may be different.
A Majority Logic Gate is used to decide which of the circuits’ outputs is the correct output. The majority gate output is 1 if two or more of the inputs of the majority gate are 1; output is 0 if two or more of the majority gate’s inputs are 0.
The Majority Logic Gate is a simple AND–OR circuit: if the inputs to the majority gate are denoted by x, y and z, then the output of the majority gate is
TMR systems should use data scrubbing—rewrite flip-flops periodically—in order to avoid accumulation of errors.
The majority gate itself could fail. Is there a way to mask that failure? In other words, who guards the guardians?
In a few TMR systems, such as the Saturn Launch Vehicle Digital Computer and functional triple modular redundancy (FTMR) systems, the voters are also triplicated. Three voters are used – one for each copy of the next stage of TMR logic. In such systems there is no single point of failure.
However, in contrast to the relatively complicated Boolean functions computed in triplicate by the TMR system, the majority gate is a simple circuit, thus its probability of failure is significantly smaller than that of each of the three circuits computing the Boolean function. In other systems there is only a single voter—If the voter fails in such a system, then the complete system will fail. However, in a good TMR system the voter is much more reliable than the other TMR components.
Triple modular redundancy in popular culture
- The three pre-cogs in Minority Report lead to a conviction even when one disagrees.
- To rule out that a single win was "a fluke", some competitions use a two out of three falls match. This isn't true TMR, however, because the three falls are not independent of each other – each competitor knows who has most falls at any point in the competition, which influences their future actions.
- In Arthur C. Clarke's science fiction novel Rendezvous with Rama, the Ramans make heavy use of triple redundancy.
See Also on BitcoinWiki
- Article about TMR with reference to TMR usage in avionics and industry
- Johnson, J. M., & Wirthlin, M. J. (2010, February). Voter insertion algorithms for FPGA designs using triple modular redundancy. In Proceedings of the 18th annual ACM/SIGDA international symposium on Field programmable gate arrays (pp. 249–258). ACM.