The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
수십 또는 수백 개의 처리 코어가 있는 다중 코어 컴퓨터에서 작업 예약은 고성능 컴퓨팅(HPC) 시스템의 핵심 기술 중 하나입니다. 많은 스케줄링 알고리즘이 제안되었음에도 불구하고 스케줄링은 다양한 스케줄링 목표를 가진 단일 컴퓨팅 노드에 할당된 매우 효과적인 작업을 실행하기 위한 과제로 남아 있습니다. 반면, 증가하는 규모와 변화하는 요구 사항에 대한 신속한 대응의 필요성은 HPC 노드의 기존 스케줄링 모델로는 충족하기 어렵습니다. 이러한 문제를 해결하기 위해 우리는 다중 코어 프로세서가 있는 단일 노드에 적용되는 새로운 적응형 스케줄링 모델을 제안합니다. 이 모델은 적응형 낙관적 제어 메커니즘을 통해 스케줄링 효율성 및 확장성 문제를 해결합니다. 이 메커니즘은 모든 코어에 해당 정보를 활용하는 데 필요한 작업 및 도구가 제공되어 조정되지 않은 방식으로 리소스를 놓고 경쟁하도록 정보를 노출합니다. 동시에 메커니즘에는 적응형 제어 기능이 탑재되어 충돌이 자주 발생할 때 실행 중인 도구의 수를 동적으로 조정할 수 있습니다. 우리는 이 스케줄링 모델을 정당화하고 합성 및 실제 HPC 워크로드에 대한 시뮬레이션 결과를 제시합니다. 여기서 제안된 모델을 널리 사용되는 두 가지 스케줄링 모델, 즉 다중 경로 모놀리식 및 2단계 스케줄링과 비교합니다. 제안된 접근 방식은 스케줄링 효율성과 확장성 측면에서 다른 모델보다 우수합니다. 우리의 결과는 적응형 낙관적 제어가 노드 수준 스케줄링 모델 및 성능의 병렬 처리에서 HPC 워크로드에 상당한 개선을 제공한다는 것을 보여줍니다.
Zhishuo ZHENG
South China University of Technology
Deyu QI
South China University of Technology
Naqin ZHOU
Guangzhou University
Xinyang WANG
South China University of Technology
Mincong YU
South China University of Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
부
Zhishuo ZHENG, Deyu QI, Naqin ZHOU, Xinyang WANG, Mincong YU, "Improving Per-Node Computing Efficiency by an Adaptive Lock-Free Scheduling Model" in IEICE TRANSACTIONS on Information,
vol. E101-D, no. 10, pp. 2423-2435, October 2018, doi: 10.1587/transinf.2018EDP7038.
Abstract: Job scheduling on many-core computers with tens or even hundreds of processing cores is one of the key technologies in High Performance Computing (HPC) systems. Despite many scheduling algorithms have been proposed, scheduling remains a challenge for executing highly effective jobs that are assigned in a single computing node with diverse scheduling objectives. On the other hand, the increasing scale and the need for rapid response to changing requirements are hard to meet with existing scheduling models in an HPC node. To address these issues, we propose a novel adaptive scheduling model that is applied to a single node with a many-core processor; this model solves the problems of scheduling efficiency and scalability through an adaptive optimistic control mechanism. This mechanism exposes information such that all the cores are provided with jobs and the tools necessary to take advantage of that information and thus compete for resources in an uncoordinated manner. At the same time, the mechanism is equipped with adaptive control, allowing it to adjust the number of running tools dynamically when frequent conflict happens. We justify this scheduling model and present the simulation results for synthetic and real-world HPC workloads, in which we compare our proposed model with two widely used scheduling models, i.e. multi-path monolithic and two-level scheduling. The proposed approach outperforms the other models in scheduling efficiency and scalability. Our results demonstrate that the adaptive optimistic control affords significant improvements for HPC workloads in the parallelism of the node-level scheduling model and performance.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDP7038/_p
부
@ARTICLE{e101-d_10_2423,
author={Zhishuo ZHENG, Deyu QI, Naqin ZHOU, Xinyang WANG, Mincong YU, },
journal={IEICE TRANSACTIONS on Information},
title={Improving Per-Node Computing Efficiency by an Adaptive Lock-Free Scheduling Model},
year={2018},
volume={E101-D},
number={10},
pages={2423-2435},
abstract={Job scheduling on many-core computers with tens or even hundreds of processing cores is one of the key technologies in High Performance Computing (HPC) systems. Despite many scheduling algorithms have been proposed, scheduling remains a challenge for executing highly effective jobs that are assigned in a single computing node with diverse scheduling objectives. On the other hand, the increasing scale and the need for rapid response to changing requirements are hard to meet with existing scheduling models in an HPC node. To address these issues, we propose a novel adaptive scheduling model that is applied to a single node with a many-core processor; this model solves the problems of scheduling efficiency and scalability through an adaptive optimistic control mechanism. This mechanism exposes information such that all the cores are provided with jobs and the tools necessary to take advantage of that information and thus compete for resources in an uncoordinated manner. At the same time, the mechanism is equipped with adaptive control, allowing it to adjust the number of running tools dynamically when frequent conflict happens. We justify this scheduling model and present the simulation results for synthetic and real-world HPC workloads, in which we compare our proposed model with two widely used scheduling models, i.e. multi-path monolithic and two-level scheduling. The proposed approach outperforms the other models in scheduling efficiency and scalability. Our results demonstrate that the adaptive optimistic control affords significant improvements for HPC workloads in the parallelism of the node-level scheduling model and performance.},
keywords={},
doi={10.1587/transinf.2018EDP7038},
ISSN={1745-1361},
month={October},}
부
TY - JOUR
TI - Improving Per-Node Computing Efficiency by an Adaptive Lock-Free Scheduling Model
T2 - IEICE TRANSACTIONS on Information
SP - 2423
EP - 2435
AU - Zhishuo ZHENG
AU - Deyu QI
AU - Naqin ZHOU
AU - Xinyang WANG
AU - Mincong YU
PY - 2018
DO - 10.1587/transinf.2018EDP7038
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2018
AB - Job scheduling on many-core computers with tens or even hundreds of processing cores is one of the key technologies in High Performance Computing (HPC) systems. Despite many scheduling algorithms have been proposed, scheduling remains a challenge for executing highly effective jobs that are assigned in a single computing node with diverse scheduling objectives. On the other hand, the increasing scale and the need for rapid response to changing requirements are hard to meet with existing scheduling models in an HPC node. To address these issues, we propose a novel adaptive scheduling model that is applied to a single node with a many-core processor; this model solves the problems of scheduling efficiency and scalability through an adaptive optimistic control mechanism. This mechanism exposes information such that all the cores are provided with jobs and the tools necessary to take advantage of that information and thus compete for resources in an uncoordinated manner. At the same time, the mechanism is equipped with adaptive control, allowing it to adjust the number of running tools dynamically when frequent conflict happens. We justify this scheduling model and present the simulation results for synthetic and real-world HPC workloads, in which we compare our proposed model with two widely used scheduling models, i.e. multi-path monolithic and two-level scheduling. The proposed approach outperforms the other models in scheduling efficiency and scalability. Our results demonstrate that the adaptive optimistic control affords significant improvements for HPC workloads in the parallelism of the node-level scheduling model and performance.
ER -