Hierarchical Dynamic Loop Self-Scheduling on Distributed-Memory Systems Using an MPI+MPI Approach

Abstract

Computationally-intensive loops are the primary source of parallelism in scientific applications. Such loops are often irregular and a balanced execution of their loop iterations is critical for achieving high performance. However, several factors may lead to an imbalanced load execution, such as problem characteristics, algorithmic, and systemic variations. Dynamic loop self-scheduling (DLS) techniques are devised to mitigate these factors, and consequently, improve application performance. On distributed-memory systems, DLS techniques can be implemented using a hierarchical master-worker execution model and are, therefore, called hierarchical DLS techniques. These techniques self-schedule loop iterations at two levels of hardware parallelism: across and within compute nodes. Hybrid programming approaches that combine the message passing interface (MPI) with open multi-processing (OpenMP) dominate the implementation of hierarchical DLS techniques. The MPI-3 standard includes the feature of sharing memory regions among MPI processes. This feature introduced the MPI+MPI approach that simplifies the implementation of parallel scientific applications. The present work designs and implements hierarchical DLS techniques by exploiting the MPI+MPI approach. Four well-known DLS techniques are considered in the evaluation proposed herein. The results indicate certain performance advantages of the proposed approach compared to the hybrid MPI+OpenMP approach.

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

“…(1) Separation between concepts and implementations: the DCA [11] and its hierarchical version [12] were motivated by the new advancements in the MPI 3.1 standard, namely MPI one-sided communication and MPI shared-memory. The following question arises: Is DCA limited to specific MPI features?…”

“…We highlight specific requirements that a DLS technique needs to fulfill to separate chunk calculation that can be distributed across all PEs and the chunk assignment that should be synchronized across all PEs. In contrast to earlier efforts [11,12], we introduce and evaluate a two-sided MPI-based implementation of DCA. This implementation applies to all existing MPI runtime libraries because they fully support two-sided MPI communication.…”

“…(2) Support for new DLS categories: Previously, DCA [11,12] only supported DLS techniques with either fixed or decreasing chunk size patterns. In this extended manuscript, we discuss how DCA supports DLS techniques that calculate fixed, decreasing, increasing, and irregular chunk size patterns.…”