Section: New Results

Data Aware Batch Scheduling

New results on the topic Data Aware Batch Scheduling are related to graph algorithm for dense k-subset detection [8], scheduling heuristic for multi-CPU multi-GPU computing platform with performance guarantee [9], machine learning for designing scheduling policies  [11] and multi-objective scheduling heuristic [13]. We detail the two last here.

• Obtaining Dynamic Scheduling Policies with Simulation and Machine Learning [11]. Dynamic scheduling of tasks in large-scale HPC platforms is normally accomplished using ad-hoc heuristics, based on task characteristics, combined with some backfilling strategy. Defining heuristics that work efficiently in different scenarios is a difficult task, specially when considering the large variety of task types and platform architectures. In this work, we present a methodology based on simulation and machine learning to obtain dynamic scheduling policies. Using simulations and a workload generation model, we can determine the characteristics of tasks that lead to a reduction in the mean slowdown of tasks in an execution queue. Modeling these characteristics using a nonlinear function and applying this function to select the next task to execute in a queue dramatically improved the mean task slowdown in synthetic workloads. When applied to real workload traces from highly different machines, these functions still resulted in important performance improvements, attesting the generalization capability of the obtained heuristics.

• A new on-line method for scheduling independent tasks [13]. We present a new method for scheduling independent tasks on a parallel machine composed of identical processors. This problem has been studied extensively for a long time with many variants. We are interested here in designing a generic algorithm in the on-line non-preemptive setting whose performance is good for various objectives. The basic idea of this algorithm is to detect some problematic tasks that are responsible for the delay of other shorter tasks. Then the former tasks are redirected to be executed in a dedicated part of the machine. We show through an extensive experimental campaign that this method is effective and in most cases is closer to some standard lower bounds than the base-line method for the problem.