Fault Tolerance Method for Mobile Cloud System Based on Hierarchical Checkpoint

Abstract

The present invention provides a mobile cloud system fault-tolerant method based on hierarchical checkpoints, including S1: dividing the checkpoints of nodes in the mobile cloud system into two-level checkpoints; S2: setting the proportion and level of the number of level-1 checkpoints ‑The proportion of the number of type 2 checkpoints; establish a model to minimize the expected total additional time overhead of checkpoints, and obtain the first failure cycle M in the mobile cloud system 1 The frequency function of the checkpoint with the least additional time overhead; according to the frequency function, obtain the placement time series of checkpoints during the execution of mobile applications by the mobile cloud system; S3: at each moment in the placement time series, set the level‑of each node Type 1 checkpoint and level-2 type checkpoint; the present invention divides the checkpoint into two levels, performs fault-tolerant configuration respectively, and performs hypergraphic modeling on the topology of the mobile application jointly executed in the mobile cloud system, reducing additional Time overhead, improving the reliability and availability of the entire mobile cloud system.

Description

technical field [0001] The invention relates to the field of mobile cloud, in particular to a fault-tolerant method for a mobile cloud system based on hierarchical checkpoints. Background technique [0002] Mobile cloud computing is considered to be one of the fastest-growing technologies. With the development of mobile cloud computing, end-user mobility has become an important feature of the contemporary Internet. Although mobile cloud computing allows mobile devices to offload computing-intensive applications to remote cloud servers for processing such as computing and storage, it enhances the performance of applications, reduces battery power consumption, and solves problems that cannot be processed in mobile devices due to resource constraints. However, problems in time management, fault-tolerant performance and load balancing are still problems that need to be optimized and solved urgently in mobile cloud computing. [0003] At present, some people have studied the mob...

AI Technical Summary

Problems solved by technology

The failure of the mobile cloud system needs to be divided into permanent failure and transient resource failure. When a permanent failure occurs on a node, the checkpoint file previously saved in the node will be lost, so that the entire mobile cloud system cannot be restarted through the checkpoint\ recovery; and as the size of the mobile cloud system expands, the size of the checkpoint file will increase accordingly, and the overhead of setting checkpoints and recovering from failures will become more and more unbearable

Images