基于Linux内核的进程检查点系统设计与实现

Design and Implementation of Process Checkpointing System Based on Linux Kernel

作为一种流行的软件容错机制,检查点与恢复技术的实现模式有两种:用户级和系统级。首先阐述了两者的区别,然后根据Linux可加载内核模块机制提出了一种基于Linux内核的进程检查点与恢复实现方法。利用Linux内核线程实现了检查点与恢复内核模块,并基于此内核模块在用户层构造了一检查点函数库,为用户提供了相应接口。用户通过组合使用这些接口可以高效地实现具体检查点与恢复算法。

As a popular software fault-tolerant mechanism, checkpoint and recovery technique can be implemented by two modes：user-level and system-level. First, the differences between the two modes were discussed. Then according to the Linux LKM （Loadable Kernel Module） mechanism,a method was proposed to design a process checkpoint and recovery system based on the Linux kernel. Checkpoint and recovery kernel module was implemented using the Linux ker- nel thread. Based on this kernel module,a checkpoint library was constructed in the user-level to provide corresponding interfaces for users. By using some selected interfaces, the particular checkpoint and recovery algorithm can he implemented effectively.