A redundant data source selection method and device

Abstract

The invention provides a redundant data source selection method and device, the method is applied to a fault-tolerant computer, the fault-tolerant computer comprises N processors, each processor is connected with a corresponding data exchange area, and each data exchange area is connected with a corresponding system I / O processing area processor. The method comprises the following steps: respectively numbering data exchange areas without priority setting in sequence, wherein the numbers comprise 1, 2... N; calibrating the current states of all the data exchange areas, wherein the current states of the data exchange areas comprise a transient fault, a permanent fault and a normal state; if it is judged that the permanent fault exists in the data exchange area, according to the serial numbers, searching the data exchange area which is closest to the serial numbers, is in a normal state and is effective and available in current data sequentially backwards to serve as the latest data exchange area; and executing the task of the period according to the data of the latest data exchange area.

Description

technical field [0001] The present application relates to the field of computers, in particular to a method and device for selecting redundant data sources. Background technique [0002] The intelligent I / O processing area under the airborne fault-tolerant computer system inevitably needs to select and use the data output by multiple redundant processors. The method that will be used is to set the priority for selection, which is for all The data exchange area corresponding to the processor is set with a fixed priority number. When the data in all areas is normal, the data in the data exchange area with high priority is used for use. When the data in the high priority data exchange area is invalid, it is valid During the use of data in the low priority data exchange area, if it is identified that the data in the high priority data exchange area meets the conditions and returns to normal, then follow the principle of priority and reselect the data that has returned to normal ...

AI Technical Summary

Problems solved by technology

[0003] This method itself has been applied more robustly. There may be a point of controversy that there is a certain imbalance in the use of all data exchange areas. Due to the recoverable characteristics in order to improve reliability, the data used is always in the lower priority. Switch between the high-priority data exchange areas, but the data in the artificially set low-priority data exchange areas cannot always be selected. Even if the data in the low-priority data exchange area is more stable, it will "never" be used

Images