Checkpointing Algorithms for Fault-Tolerant Execution of Large-Scale Distributed Applications in Cloud

摘要: Cloud computing provides infinite resources and a suitable environment for the execution of large scale applications. However, it is also susceptible to frequent failures which can affect users as well service providers adversely. Therefore, fault tolerance techniques are necessary reliable applications in cloud. This work presents checkpointing based protocols two types distributed The first kind Bags Tasks (BoT) where an application comprises set independent tasks that do not communicate with each other during execution. Hence, uncoordinated algorithm proposed BoT Subsequently, we consider composed multiple dependent on due inter-task message passing. An logging protocol presented this type utilize storage at edge switches data center reduce bandwidth consumption saving checkpoints logs. Simulation results have demonstrated provide increased rate successful recoveries from cause lower resource overhead than contemporary related schemes.