Scheduling Scientific Workloads in Private Cloud: Problems and
Approaches

D 2017

Scheduling Scientific Workloads in Private Cloud: Problems and Approaches

KLUSÁČEK, Dalibor, Boris PARÁK, Gabriela PODOLNÍKOVÁ a András ÜRGE

Základní údaje

Originální název

Scheduling Scientific Workloads in Private Cloud: Problems and Approaches

Autoři

KLUSÁČEK, Dalibor (203 Česká republika), Boris PARÁK (703 Slovensko), Gabriela PODOLNÍKOVÁ (203 Česká republika, garant, domácí) a András ÜRGE (703 Slovensko, domácí)

Vydání

New York, 10th International Conference on Utility and Cloud Computing (UCC 2017), od s. 9-18, 10 s. 2017

Nakladatel

ACM

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10201 Computer sciences, information science, bioinformatics

Stát vydavatele

Spojené státy

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

elektronická verze "online"

Kód RIV

RIV/00216224:14330/17:00118585

Organizační jednotka

Fakulta informatiky

ISBN

978-1-4503-5149-2

DOI

http://dx.doi.org/10.1145/3147213.3147223

UT WoS

000568256400005

Klíčová slova anglicky

Private Cloud; Fairness; Fair-Sharing; Scheduling; OpenNebula

Štítky

firank_B

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 5. 11. 2021 15:02, RNDr. Pavel Šmerk, Ph.D.

Anotace

V originále

Public cloud providers are using the "pay-per-use" model when providing their resources to customers. Among other advantages, it allows the provider to react to changing demands, e.g., by modifying prices or by extending its physical capacities using the profit obtained. In this paper we deal with a completely different model. We describe a private scientific cloud where resources are provided to researchers for free. As we demonstrate, the "absence of money" means that the system must employ other mechanisms to guarantee reasonable performance and utilization. Especially, the problem of guaranteeing user-to-user fairness represents a major issue. Moreover, since there is no financial burden related to the use of cloud infrastructure, many resources can be wasted by long running idle virtual machines (VM) that their users no longer need. This leads to underutilization and resource fragmentation. This paper discusses these problems using real-life data from the CERIT Scientific Cloud and proposes several techniques to guarantee fair and efficient use of system resources. Furthermore, we present a prototype of a new experimental OpenNebula-compatible VM scheduler which was designed as a replacement for the default scheduler provided in OpenNebula distribution. Unlike the default scheduler, our new scheduler provides complex fair-sharing mechanisms as well as modular and easy-to-extend architecture to enable further development of advanced VM scheduling policies.

Citovat

KLUSÁČEK, Dalibor, Boris PARÁK, Gabriela PODOLNÍKOVÁ a András ÜRGE. Scheduling Scientific Workloads in Private Cloud: Problems and Approaches. Online. In 10th International Conference on Utility and Cloud Computing (UCC 2017). New York: ACM, 2017, s. 9-18. ISBN 978-1-4503-5149-2. Dostupné z: https://dx.doi.org/10.1145/3147213.3147223.

@inproceedings{1768262,
   author = {Klusáček, Dalibor and Parák, Boris and Podolníková, Gabriela and Ürge, András},
   address = {New York},
   booktitle = {10th International Conference on Utility and Cloud Computing (UCC 2017)},
   doi = {http://dx.doi.org/10.1145/3147213.3147223},
   keywords = {Private Cloud; Fairness; Fair-Sharing; Scheduling; OpenNebula},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {New York},
   isbn = {978-1-4503-5149-2},
   pages = {9-18},
   publisher = {ACM},
   title = {Scheduling Scientific Workloads in Private Cloud: Problems and Approaches},
   year = {2017}
}

TY  - JOUR
ID  - 1768262
AU  - Klusáček, Dalibor - Parák, Boris - Podolníková, Gabriela - Ürge, András
PY  - 2017
TI  - Scheduling Scientific Workloads in Private Cloud: Problems and Approaches
PB  - ACM
CY  - New York
SN  - 9781450351492
KW  - Private Cloud
KW  - Fairness
KW  - Fair-Sharing
KW  - Scheduling
KW  - OpenNebula
N2  - Public cloud providers are using the "pay-per-use" model when providing their resources to customers. Among other advantages, it allows the provider to react to changing demands, e.g., by modifying prices or by extending its physical capacities using the profit obtained. In this paper we deal with a completely different model. We describe a private scientific cloud where resources are provided to researchers for free. As we demonstrate, the "absence of money" means that the system must employ other mechanisms to guarantee reasonable performance and utilization. Especially, the problem of guaranteeing user-to-user fairness represents a major issue. Moreover, since there is no financial burden related to the use of cloud infrastructure, many resources can be wasted by long running idle virtual machines (VM) that their users no longer need. This leads to underutilization and resource fragmentation. This paper discusses these problems using real-life data from the CERIT Scientific Cloud and proposes several techniques to guarantee fair and efficient use of system resources. Furthermore, we present a prototype of a new experimental OpenNebula-compatible VM scheduler which was designed as a replacement for the default scheduler provided in OpenNebula distribution. Unlike the default scheduler, our new scheduler provides complex fair-sharing mechanisms as well as modular and easy-to-extend architecture to enable further development of advanced VM scheduling policies.
ER  -

KLUSÁČEK, Dalibor, Boris PARÁK, Gabriela PODOLNÍKOVÁ a András ÜRGE. Scheduling Scientific Workloads in Private Cloud: Problems and Approaches. Online. In \textit{10th International Conference on Utility and Cloud Computing (UCC 2017)}. New York: ACM, 2017, s.~9-18. ISBN~978-1-4503-5149-2. Dostupné z: https://dx.doi.org/10.1145/3147213.3147223.

Podrobný výpis o publikaci