设置进程检查点的嵌入式容错系统设计

针对嵌入式Linux系统的特点,通过设置检查点(checkpoint)实现ARM平台进程级容错。在检查点工作时,通过/proc文件系统与内核进行交互,实时地获取与进程有关的PID、CPU状态以及内存信息,并保存在存储介质中。当进程出现故障后,将上述与进程有关的状态信息进行恢复,从而实现进程级容错。实验表明,该进程级容错系统有较好的容错能力,极大地缩短了进程恢复的时间。

Design of Embedded Fault Tolerant System Based on Process Checkpoint Settings

According to the characteristics of embedded Linux system, the process-lever fault tolerant of ARM platform is realized through setting checkpoint. During the working,the checkpoint realizes the interactions with the kernel through the proc file system. At the same time, the checkpoint gets the PID,CPU status and memory information about the process, and stores the information in the storage medium. When the process is failed, the system resumes the above information. The experiments indicate this process-level fault tolerant system achieves fault tolerance and reduces the time of resuming process.