This paper models the task assignment based on guessing size (TAGS) job allocation algorithm using Markovian processing algebra; PEPA. It aims to analyse performance and energy consumption. The working environment is assumed to be heterogeneous, and the job size distribution is assumed to be a two phase hyper-exponential. Furthermore, the queues are bounded. A two nodes system is implemented with exponentially distributed incoming tasks. We analysed the performance metrics and energy consumption under different arrival rates. We found TAGS can perform well and improve performance, although it increases total energy consumption. Finally, we calculated the energy per job to evaluate TAGS in a heterogeneous environment, and demonstrated that TAGS reduces energy consumption per job when the system is under a heavy load.

本文采用马尔可夫处理代数，基于猜测大小（TAGS）作业分配算法对任务分配进行建模。PEPA。它旨在分析​​性能和能耗。假定工作环境是异构的，并且作业大小分布假定为两相超指数。此外，队列是有界的。两节点系统以指数分布的传入任务实现。我们分析了不同到达率下的性能指标和能耗。我们发现TAGS尽管会增加总能耗，但可以很好地执行并提高性能。最后，我们计算了异构环境中TAGS的每项作业能耗，并证明了当系统处于重负载时TAGS可以降低每项作业的能耗。