Master-checker
|
|
A master-checker is a hardware-supported fault tolerance method for multiprocessor systems.
Two processors calculate in parallel the same functions to increase the probability that the result is exact.
In technical terms
The checker-CPU is synchronised at clock level with the master-CPU and processes the very same programs as the master. Whenever the master-CPU generates an output, the checker-CPU compares this output to its own calculation and in the event of a difference raises a warning.
Advantages
Advantages to the master-checker system is in the legitimacy of the answer. Processors calculating the same algorithm insure that the answer is correct before passing it on to the application requesting the algorithm being completed. If an error occurs, the process can be recalculated and/or a message sent to the application, operating system, and/or user. A recurrance of discrepencies between the two processor could indicates a flaw in the software, hardware problems, or timing issues between the clock, CPUs, and/or system memory.
Disadvantages
Multiple processors calculating the same algorith wastes time and energy. If the master-CPU is correct 95% or more of the time, the power and time used by the checker-CPU to check the master-CPU is wasted. Depending on the value of a correct answer a high percentage of the time, a checker-CPU may or may not be warranted. Also, the checker-CPU could be used to calculate something else in the same algorithm increasing the speed and processing output of the CPU system.