Number of faults a system can withstand without repairs
arXiv:math/9409220
Abstract
We consider the following scheduling problem. A system is composed of processors drawn from a pool of . The processors can become faulty while in operation and faulty processors never recover. A report is issued whenever a fault occurs. This report states only the existence of a fault, but does not indicate its location. Based on this report, the scheduler can reconfigure the system and choose another set of processors. The system operates satisfactorily as long as at most of the selected processors are faulty. We exhibit a scheduling strategy allowing the system to operate satisfactorily until approximately faults are reported in the worst case. Our precise bound is tight.
10 pages