Job simulation for large-scale PBS based clusters with the Maui Scheduler

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27740%2F18%3A10239712" target="_blank" >RIV/61989100:27740/18:10239712 - isvavai.cz</a>

Výsledek na webu

<a href="http://www.iadisportal.org/digital-library/iadis-international-conference-theory-and-practice-in-modern-computing-2018-part-of-mccsis-2018" target="_blank" >http://www.iadisportal.org/digital-library/iadis-international-conference-theory-and-practice-in-modern-computing-2018-part-of-mccsis-2018</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Job simulation for large-scale PBS based clusters with the Maui Scheduler

Popis výsledku v původním jazyce

For large-scale High Performance Computing centers with a wide range of different projects and heterogeneous infrastructures, efficiency is an important consideration. Understanding how compute jobs are scheduled is necessary for improving the job scheduling strategies in order to optimize cluster utilization and job wait times. This increases the importance of a reliable simulation capability, which in turn requires accuracy and comparability with historic workloads from the cluster. Not all job schedulers have a simulation capability, including the Portable Batch System (PBS) resource manager. Hence, PBS based centers have no direct way to simulate changes and optimizations before they are applied to the production system. We propose and discuss how to run job simulations for large-scale PBS based clusters with the Maui Scheduler. For validation purposes, we use historic workloads collected at the IT4Innovations supercomputing center, and demonstrate the viability of our approach. We are confident that our approach is also transferable to enable job simulations for other computing centers using PBS.

Název v anglickém jazyce

Job simulation for large-scale PBS based clusters with the Maui Scheduler

Popis výsledku anglicky

For large-scale High Performance Computing centers with a wide range of different projects and heterogeneous infrastructures, efficiency is an important consideration. Understanding how compute jobs are scheduled is necessary for improving the job scheduling strategies in order to optimize cluster utilization and job wait times. This increases the importance of a reliable simulation capability, which in turn requires accuracy and comparability with historic workloads from the cluster. Not all job schedulers have a simulation capability, including the Portable Batch System (PBS) resource manager. Hence, PBS based centers have no direct way to simulate changes and optimizations before they are applied to the production system. We propose and discuss how to run job simulations for large-scale PBS based clusters with the Maui Scheduler. For validation purposes, we use historic workloads collected at the IT4Innovations supercomputing center, and demonstrate the viability of our approach. We are confident that our approach is also transferable to enable job simulations for other computing centers using PBS.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

MCCSIS 2018 - Multi Conference on Computer Science and Information Systems; Proceedings of the International Conferences on Big Data Analytics, Data Mining and Computational Intelligence 2018, Theory and Practice in Modern Computing 2018 and Connected Smart Cities 2018

ISBN

978-989-8533-80-7

ISSN

—

e-ISSN

neuvedeno

Počet stran výsledku

9

Strana od-do

137-145

Název nakladatele

IADIS

Místo vydání

Lisabon

Místo konání akce

Madrid

Datum konání akce

17. 7. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—