Task Allocation and Checkpointing in Distributed Computer Systems

随着基于消息的分布式并行计算技术的不断发展， 计算系统，与构建相关的困难 并行算法并将其映射到并行体系结构上， 重要性增长。 目前，算法开发和 编程过程通常包括以下内容的显式映射： 算法子任务到处理器。 这个问题通常是由 手动地并且经常导致低效的计算， 通信负载平衡，这反过来又导致整体 降低了可能获得的加速比。 当分布式体系结构 在处理器和/或通信链路受到 失败，但必须继续以合理的速度处理， 剩余的计算资源。 这就要求流程和数据 检查点必须有效地进行，并在检测和定位 故障时，重新配置系统，重新分配任务。 这 工作集中于检查点和任务重新分配问题。 该研究计划旨在调查方法， 将任务分配给分布式计算机系统中的处理器， 可靠性特征已被表征。 目标是 确保有效利用剩余的分布式资源， 故障（处理器或链路）发生在 分布式检查点和快速任务重新分配方案。

With the increasing availability of message based distributed parallel computing systems, the difficulties associated with constructing parallel algorithms and mapping them onto parallel architectures have grown in importance. Currently, the algorithm development and programming process typically includes the explicit mapping of algorithm subtasks to processors. This problem is typically dealt with manually and often results in inefficient computation and communication load balancing which in turn leads to an overall reduction in the speedups potentially obtainable. The problem is exacerbated when distributed architectures are considered where processors and/or communications links are subject to failure, yet must continue processing at reasonable rates with the remaining computing resources. This requires that process and data checkpointing be done effectively and that, on detection and location of failures, the system be reconfigured and tasks reallocated. This work focuses on the checkpointing and task reallocation problems. This research program is aimed at investigating approaches to allocating tasks to processors in distributed computer systems whose reliability characteristics have been characterized. The goal is to ensure effective use of the remaining distributed resources after failure (processor or link) has occurred through the development of distributed checkpointing and fast task reallocation schemes.