Abstract

In high performance computing (HPC) job schedulers usually divide resources of computing nodes into slots. Each slot can be assigned to execute only a single job from the queue. In some cases, jobs do not fully utilize all available resources from the slot which leads to internal fragmentation, wasted resources and to an increase of queue wait time. In this paper, we propose fair resource allocation strategies that can be applied in job schedulers for resource allocation. We cover such resources as CPU time, residential memory and network bandwidth.

Original languageEnglish
Pages (from-to)740-751
Number of pages12
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11622
DOIs
Publication statusPublished - 1 Jul 2019
Event19th International Conference on Computational Science and Its Applications, ICCSA 2019 - Saint Petersburg
Duration: 1 Jul 20194 Jul 2019

Fingerprint

Resource Allocation
Resource allocation
Workload
High Performance
Resources
Computing
Scheduler
Program processors
Queue
Bandwidth
Data storage equipment
CPU Time
Fragmentation
Divides
Cover
Internal
Vertex of a graph

Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

@article{253c5a3a493c4ce8bb2d62bcc441705f,
title = "Fair Resource Allocation for Running HPC Workloads Simultaneously",
abstract = "In high performance computing (HPC) job schedulers usually divide resources of computing nodes into slots. Each slot can be assigned to execute only a single job from the queue. In some cases, jobs do not fully utilize all available resources from the slot which leads to internal fragmentation, wasted resources and to an increase of queue wait time. In this paper, we propose fair resource allocation strategies that can be applied in job schedulers for resource allocation. We cover such resources as CPU time, residential memory and network bandwidth.",
keywords = "Fair resource allocation, High performance computing, Scheduling",
author = "Ruslan Kuchumov and Vladimir Korkhov",
year = "2019",
month = "7",
day = "1",
doi = "10.1007/978-3-030-24305-0_55",
language = "English",
volume = "11622",
pages = "740--751",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer",

}

TY - JOUR

T1 - Fair Resource Allocation for Running HPC Workloads Simultaneously

AU - Kuchumov, Ruslan

AU - Korkhov, Vladimir

PY - 2019/7/1

Y1 - 2019/7/1

N2 - In high performance computing (HPC) job schedulers usually divide resources of computing nodes into slots. Each slot can be assigned to execute only a single job from the queue. In some cases, jobs do not fully utilize all available resources from the slot which leads to internal fragmentation, wasted resources and to an increase of queue wait time. In this paper, we propose fair resource allocation strategies that can be applied in job schedulers for resource allocation. We cover such resources as CPU time, residential memory and network bandwidth.

AB - In high performance computing (HPC) job schedulers usually divide resources of computing nodes into slots. Each slot can be assigned to execute only a single job from the queue. In some cases, jobs do not fully utilize all available resources from the slot which leads to internal fragmentation, wasted resources and to an increase of queue wait time. In this paper, we propose fair resource allocation strategies that can be applied in job schedulers for resource allocation. We cover such resources as CPU time, residential memory and network bandwidth.

KW - Fair resource allocation

KW - High performance computing

KW - Scheduling

UR - http://www.scopus.com/inward/record.url?scp=85068620987&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-24305-0_55

DO - 10.1007/978-3-030-24305-0_55

M3 - Article

AN - SCOPUS:85068620987

VL - 11622

SP - 740

EP - 751

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -