Abstract

The fault-tolerant ability of an object-based storage system is an important factor that affects the reliability and usability of the entire system. Existing fault-tolerant methods either increase procurement cost, or impede performance. In this paper, we propose an I/O fault-tolerant method based on an application-level agent, which can perform fault-tolerance swiftly and efficiently at runtime by constructing the fault-tolerant agent’s execution engine. We consider a “disk full” exception as an example to further detail the implementation of this method. Experimental results confirm the method’s effectiveness.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call