An Investigation into Soft Error Detection Efficiency at Operating System Level

الملخص EN

Electronic equipment operating in harsh environments such as space is subjected to a range of threats.

The most important of these is radiation that gives rise to permanent and transient errors on microelectronic components.

The occurrence rate of transient errors is significantly more than permanent errors.

The transient errors, or soft errors, emerge in two formats: control flow errors (CFEs) and data errors.

Valuable research results have already appeared in literature at hardware and software levels for their alleviation.

However, there is the basic assumption behind these works that the operating system is reliable and the focus is on other system levels.

In this paper, we investigate the effects of soft errors on the operating system components and compare their vulnerability with that of application level components.

Results show that soft errors in operating system components affect both operating system and application level components.

Therefore, by providing endurance to operating system level components against soft errors, both operating system and application level components gain tolerance.