A survey on software checkpointing and mobility techniques in distributed systems

Abstract

SUMMARYThis paper has two purposes. First, it shows that existing solutions employing checkpointing and mobility in distributed applications, fix, at design time, the types and the techniques of checkpointing and mobility to be employed at runtime. Second, it provides policies allowing the dynamic selection of checkpointing and mobility techniques according to the execution environment. For that, this paper presents checkpointing and mobility techniques to evaluate them in order to identify their advantages and their shortcomings, and then determine the appropriate execution conditions in which a specific mobility technique becomes beneficial. This investigation will allow in assisting adaptation plans' generation and promoting future research in the area of self-adaptivity of distributed applications.

Checkpointing Models for Tasks with Widely Different Processing Times

“…Research advances on checkpointing have also been 'checkpointed' by surveys at regular intervals. Worth mentioning are the surveys by Chandy [4], Nicola [22], Elnozahy et al [12] and Marzouk and Jmaiel [20]. The Elnozahy et.…”

V2PFQL: A proactive fault tolerance approach for cloud‐hosted applications in cloud computing environment

“…It can be applied as both auxiliary and stand‐alone fault tolerant. To reduce overhead, the frequency of saving checkpoints can be adjusted based on the system fault rates [16]. If one of the tasks' execution resources fails, the task is moved to another similar resource using the job migration technique.…”