System Software Support for Resilience
摘要
Resilience of a system is not property any longer, it is a continuous process to prove that system is capable and operates as required… Our generalized algorithm of fault tolerance (GAFT) became a framework for pursuing resilience. GAFT might be efficiently implemented by hardware and system software holistically, pursuing all steps in concert—as one entity, together. This chapter presents essential features of software to support resilience: procedure of testingTesting required to deal with different duration and character of fault, a monitoring of hardware configuration even in presence of hardware faults. We also present a scheduling of testingTesting to provide required level of system integrity during its functioning. In the structure of operating system, we introduce a scheme and algorithm of control of hardware reconfiguration—so-called hardware monitor.