Supercomputer Workload Characterization

这项工作将提供一个工作量的特点， 主要工作类别的分类、工作混合比率 参数更改，队列的组成作为作业大小的函数 和CPU限制，以及资源需求和 应用算法 我们的目标是尽量减少挂钟 通过最佳队列控制，最大限度地减少空闲时间， 在真实的内存中保持适当的作业组合， 在一般的组合中，不惩罚所有 过程，包括最大的过程。

The work will provide a workload characterization based on categorization of major job classes, rates at which the job mix parameters change, composition of queues as a function of job size and CPU limits, and a correlation between resource demands and application algorithms. The goals are to minimize the wall clock expansion factor through optimal queue control, minimize idle time by keeping the appropriate job mix in real memory, and process large resource demanding jobs in the general mix, without penalizing all processes, including the largest ones.