The grand CRU challenge

Marcel Blöcher; Malte Viering; Stefan Schmid; Patrick Eugster

doi:10.1145/3094405.3094407

The grand CRU challenge

Marcel Blöcher, Malte Viering, Stefan Schmid, Patrick Eugster

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

Abstract

One of the main objectives of any cluster management system is the maximization of cluster resource utilization (CRU). In this paper, we argue that there is a dilemma underlying the challenge of maximizing CRU, as soon as network resources enter the picture. In contrast to local resources which can be handled in a more isolated fashion, global network resources are namely shared, and their allocation is intertwined with that of local resources. For effective resource management, either applications thus have to learn more about the infrastructure, or the resource manager has to understand application semantics - both options violate the separation of applications from the underlying infrastructure strived for by resource managers. This paper makes the case for a resource management system that addresses the dilemma, and presents first ideas.

Original language	English
Title of host publication	HotConNet 2017 - Proceedings of the 2017 Workshop on Hot Topics in Container Networking and Networked Systems, Part of SIGCOMM 2017
Number of pages	5
Publisher	Association for Computing Machinery
Publication date	11 Aug 2017
Pages	7-11
ISBN (Electronic)	9781450350587
DOIs	https://doi.org/10.1145/3094405.3094407
Publication status	Published - 11 Aug 2017
Event	1st ACM SIGCOMM Hot Topics on Container Networking and Networked Systems Workshop, HotConNet 2017 - Los Angeles, United States Duration: 25 Aug 2017 → …

Conference

Conference	1st ACM SIGCOMM Hot Topics on Container Networking and Networked Systems Workshop, HotConNet 2017
Country/Territory	United States
City	Los Angeles
Period	25/08/2017 → …
Sponsor	ACM SigComm

Keywords

Cloud computing
Cluster scheduling
Resource manager architecture

Access to Document

10.1145/3094405.3094407

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@inproceedings{9e50f1189c3446a8afd3e373eb92092c,

title = "The grand CRU challenge",

abstract = "One of the main objectives of any cluster management system is the maximization of cluster resource utilization (CRU). In this paper, we argue that there is a dilemma underlying the challenge of maximizing CRU, as soon as network resources enter the picture. In contrast to local resources which can be handled in a more isolated fashion, global network resources are namely shared, and their allocation is intertwined with that of local resources. For effective resource management, either applications thus have to learn more about the infrastructure, or the resource manager has to understand application semantics - both options violate the separation of applications from the underlying infrastructure strived for by resource managers. This paper makes the case for a resource management system that addresses the dilemma, and presents first ideas.",

keywords = "Cloud computing, Cluster scheduling, Resource manager architecture",

author = "Marcel Bl{\"o}cher and Malte Viering and Stefan Schmid and Patrick Eugster",

year = "2017",

month = aug,

day = "11",

doi = "10.1145/3094405.3094407",

language = "English",

pages = "7--11",

booktitle = "HotConNet 2017 - Proceedings of the 2017 Workshop on Hot Topics in Container Networking and Networked Systems, Part of SIGCOMM 2017",

publisher = "Association for Computing Machinery",

address = "United States",

note = "1st ACM SIGCOMM Hot Topics on Container Networking and Networked Systems Workshop, HotConNet 2017 ; Conference date: 25-08-2017",

}

Blöcher, M, Viering, M, Schmid, S & Eugster, P 2017, The grand CRU challenge. in HotConNet 2017 - Proceedings of the 2017 Workshop on Hot Topics in Container Networking and Networked Systems, Part of SIGCOMM 2017. Association for Computing Machinery, pp. 7-11, 1st ACM SIGCOMM Hot Topics on Container Networking and Networked Systems Workshop, HotConNet 2017, Los Angeles, United States, 25/08/2017. https://doi.org/10.1145/3094405.3094407

The grand CRU challenge. / Blöcher, Marcel; Viering, Malte; Schmid, Stefan et al.
HotConNet 2017 - Proceedings of the 2017 Workshop on Hot Topics in Container Networking and Networked Systems, Part of SIGCOMM 2017. Association for Computing Machinery, 2017. p. 7-11.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - The grand CRU challenge

AU - Blöcher, Marcel

AU - Viering, Malte

AU - Schmid, Stefan

AU - Eugster, Patrick

PY - 2017/8/11

Y1 - 2017/8/11

N2 - One of the main objectives of any cluster management system is the maximization of cluster resource utilization (CRU). In this paper, we argue that there is a dilemma underlying the challenge of maximizing CRU, as soon as network resources enter the picture. In contrast to local resources which can be handled in a more isolated fashion, global network resources are namely shared, and their allocation is intertwined with that of local resources. For effective resource management, either applications thus have to learn more about the infrastructure, or the resource manager has to understand application semantics - both options violate the separation of applications from the underlying infrastructure strived for by resource managers. This paper makes the case for a resource management system that addresses the dilemma, and presents first ideas.

AB - One of the main objectives of any cluster management system is the maximization of cluster resource utilization (CRU). In this paper, we argue that there is a dilemma underlying the challenge of maximizing CRU, as soon as network resources enter the picture. In contrast to local resources which can be handled in a more isolated fashion, global network resources are namely shared, and their allocation is intertwined with that of local resources. For effective resource management, either applications thus have to learn more about the infrastructure, or the resource manager has to understand application semantics - both options violate the separation of applications from the underlying infrastructure strived for by resource managers. This paper makes the case for a resource management system that addresses the dilemma, and presents first ideas.

KW - Cloud computing

KW - Cluster scheduling

KW - Resource manager architecture

UR - http://www.scopus.com/inward/record.url?scp=85030786000&partnerID=8YFLogxK

U2 - 10.1145/3094405.3094407

DO - 10.1145/3094405.3094407

M3 - Article in proceeding

AN - SCOPUS:85030786000

SP - 7

EP - 11

BT - HotConNet 2017 - Proceedings of the 2017 Workshop on Hot Topics in Container Networking and Networked Systems, Part of SIGCOMM 2017

PB - Association for Computing Machinery

T2 - 1st ACM SIGCOMM Hot Topics on Container Networking and Networked Systems Workshop, HotConNet 2017

Y2 - 25 August 2017

ER -

The grand CRU challenge

Abstract

Conference

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Cite this