SimAS: A simulation‐assisted approach for the scheduling algorithm selection under perturbations

Abstract

Many scientific applications consist of large and computationally-intensive loops, such as N-body, Monte Carlo, and computational fluid dynamics These loops contain computationally-intensive operations, resulting in heavy loop bodies.Dynamic loop self-scheduling (DLS) techniques are used to parallelize and to balance the load during the execution of such applications. Load imbalance arises from variations in the loop iteration (or tasks) execution times, caused by problem, algorithmic, or systemic characteristics. The variations in systemic characteristics are referred to as perturbations, and can be caused by other applications or processes that share the same resources, or a temporary system fault or malfunction and include, decreased delivered computational speed, reduced available network bandwidth, or larger network latencies. DLS achieves a balanced load execution of scientific applications on high-performance computing (HPC) systems. Therefore, the selection of the most efficient DLS technique is critical to achieve the best application performance. The following question motivates this work: "Given an application, an HPC system, and their characteristics and interplay, which DLS technique will achieve improved performance under unpredictable perturbations?" Existing studies focus on variations in the delivered computational speed only as the source of perturbations in the system. However, perturbations in available network bandwidth or latency are inevitable on production HPC systems. Also, scheduling solutions based on optimization techniques, e.g., evolutionary algorithms, can not adapt to perturbations during execution. The alternative of using machine learning for DLS selection requires training and learning either prior to execution or during previous time-steps in time-stepping applications. A Simulator-assisted scheduling (SimAS ) is introduced as a new control-theoretic-inspired approach to dynamically select DLS techniques that improve the performance of applications executing on heterogeneous HPC systems under perturbations. The present work examines the performance of seven applications on a heterogeneous system under all the above system perturbations. SimAS is evaluated as a proof of concept using native and simulative experiments. The performance results confirm the original hypothesis that no single DLS technique can deliver the absolute best performance in all scenarios, whereas the SimAS -based DLS selection resulted in improved application performance in most experiments. SS and AWF-C are the most efficient in pea-es (b) Application performance under various perturbations using different scheduling techniques.designed for time-stepping applications. It improves WF by adapting the relative weights of PEs during execution by monitoring their performance in each time-step. AWF-B relieves the time-stepping requirement in AWF, and measures the performance after each batch to update the PE weights. AWF-C is similar to AWF-B where weight updates are performed upon the completion of each chunk, ...

A Distributed Chunk Calculation Approach for Self-scheduling of Parallel Applications on Distributed-memory Systems

“…LB4MPI [14,23] is an extension of the LB tool [13] that includes certain bug fixes and additional DLS techniques. Both LB and LB4MPI employ a master-worker execution in which the master is a central entity that performs both of chunk calculation and the chunk assignment operations.…”

“…LB4MPI 1 [14,23] is a recent MPI-based library for loop scheduling and dynamic load balancing. LB4MPI extends the LB tool [13] by including certain bug fixes and additional DLS techniques.…”