Exploring potential for non-disruptive vertical auto scaling and resource estimation in kubernetes

Gourav Rattihalli; Madhusudhan Govindaraju; Hui Lu; Devesh Tiwari

doi:10.1109/CLOUD.2019.00018

Exploring potential for non-disruptive vertical auto scaling and resource estimation in kubernetes

Gourav Rattihalli
, Madhusudhan Govindaraju
, Hui Lu
, Devesh Tiwari

International Stdnts Schr Srvc

Binghamton University

State University of New York Binghamton University
Northeastern University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

93 Scopus citations

Abstract

Cloud platforms typically require users to provide resource requirements for applications so that resource managers can schedule containers with adequate allocations. However, the requirements for container resources often depend on numerous factors such as application input parameters, optimization flags, input files, and attributes that are specified for each run. So, it is complex for users to estimate the resource requirements for a given container accurately, leading to resource over-estimation that negatively affects overall utilization. We have designed a Resource Utilization Based Autoscaling System (RUBAS) that can dynamically adjust the allocation of containers running in a Kubernetes cluster. RUBAS improves upon the Kubernetes Vertical Pod Autoscaler (VPA) system non-disruptively by incorporating container migration. Our experiments use multiple scientific benchmarks. We analyze the allocation pattern of RUBAS with Kubernetes VPA. We compare the performance of container migration for in-place and remote node migration and we evaluate the overhead in RUBAS. Our results show that compared to Kubernetes VPA, RUBAS improves the CPU and memory utilization of the cluster by 10% and reduces the runtime by 15% with an overhead for each application ranging from 5% to 20%.

Original language	English
Title of host publication	Proceedings - 2019 IEEE International Conference on Cloud Computing, CLOUD 2019 - Part of the 2019 IEEE World Congress on Services
Editors	Elisa Bertino, Carl K. Chang, Peter Chen, Ernesto Damiani, Michael Goul, Katsunori Oyama
Publisher	IEEE Computer Society
Pages	33-40
Number of pages	8
ISBN (Electronic)	9781728127057
DOIs	https://doi.org/10.1109/CLOUD.2019.00018
State	Published - Jul 2019
Event	12th IEEE International Conference on Cloud Computing, CLOUD 2019 - Milan, Italy Duration: Jul 8 2019 → Jul 13 2019

Publication series

Name	IEEE International Conference on Cloud Computing, CLOUD
Volume	2019-July

Conference

Conference	12th IEEE International Conference on Cloud Computing, CLOUD 2019
Country/Territory	Italy
City	Milan
Period	07/8/19 → 07/13/19

Keywords

Container migration
Docker
Docker Swarm
Kubernetes
VPA
Vertical Scaling

Access to Document

10.1109/CLOUD.2019.00018

Cite this

Rattihalli, G., Govindaraju, M., Lu, H., & Tiwari, D. (2019). Exploring potential for non-disruptive vertical auto scaling and resource estimation in kubernetes. In E. Bertino, C. K. Chang, P. Chen, E. Damiani, M. Goul, & K. Oyama (Eds.), Proceedings - 2019 IEEE International Conference on Cloud Computing, CLOUD 2019 - Part of the 2019 IEEE World Congress on Services (pp. 33-40). Article 8814504 (IEEE International Conference on Cloud Computing, CLOUD; Vol. 2019-July). IEEE Computer Society. https://doi.org/10.1109/CLOUD.2019.00018

Rattihalli, Gourav ; Govindaraju, Madhusudhan ; Lu, Hui et al. / Exploring potential for non-disruptive vertical auto scaling and resource estimation in kubernetes. Proceedings - 2019 IEEE International Conference on Cloud Computing, CLOUD 2019 - Part of the 2019 IEEE World Congress on Services. editor / Elisa Bertino ; Carl K. Chang ; Peter Chen ; Ernesto Damiani ; Michael Goul ; Katsunori Oyama. IEEE Computer Society, 2019. pp. 33-40 (IEEE International Conference on Cloud Computing, CLOUD).

@inproceedings{126a48979c79483a9975789aac7474c2,

title = "Exploring potential for non-disruptive vertical auto scaling and resource estimation in kubernetes",

abstract = "Cloud platforms typically require users to provide resource requirements for applications so that resource managers can schedule containers with adequate allocations. However, the requirements for container resources often depend on numerous factors such as application input parameters, optimization flags, input files, and attributes that are specified for each run. So, it is complex for users to estimate the resource requirements for a given container accurately, leading to resource over-estimation that negatively affects overall utilization. We have designed a Resource Utilization Based Autoscaling System (RUBAS) that can dynamically adjust the allocation of containers running in a Kubernetes cluster. RUBAS improves upon the Kubernetes Vertical Pod Autoscaler (VPA) system non-disruptively by incorporating container migration. Our experiments use multiple scientific benchmarks. We analyze the allocation pattern of RUBAS with Kubernetes VPA. We compare the performance of container migration for in-place and remote node migration and we evaluate the overhead in RUBAS. Our results show that compared to Kubernetes VPA, RUBAS improves the CPU and memory utilization of the cluster by 10\% and reduces the runtime by 15\% with an overhead for each application ranging from 5\% to 20\%.",

keywords = "Container migration, Docker, Docker Swarm, Kubernetes, VPA, Vertical Scaling",

author = "Gourav Rattihalli and Madhusudhan Govindaraju and Hui Lu and Devesh Tiwari",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 12th IEEE International Conference on Cloud Computing, CLOUD 2019 ; Conference date: 08-07-2019 Through 13-07-2019",

year = "2019",

month = jul,

doi = "10.1109/CLOUD.2019.00018",

language = "English",

series = "IEEE International Conference on Cloud Computing, CLOUD",

publisher = "IEEE Computer Society",

pages = "33--40",

editor = "Elisa Bertino and Chang, \{Carl K.\} and Peter Chen and Ernesto Damiani and Michael Goul and Katsunori Oyama",

booktitle = "Proceedings - 2019 IEEE International Conference on Cloud Computing, CLOUD 2019 - Part of the 2019 IEEE World Congress on Services",

}

Rattihalli, G, Govindaraju, M, Lu, H & Tiwari, D 2019, Exploring potential for non-disruptive vertical auto scaling and resource estimation in kubernetes. in E Bertino, CK Chang, P Chen, E Damiani, M Goul & K Oyama (eds), Proceedings - 2019 IEEE International Conference on Cloud Computing, CLOUD 2019 - Part of the 2019 IEEE World Congress on Services., 8814504, IEEE International Conference on Cloud Computing, CLOUD, vol. 2019-July, IEEE Computer Society, pp. 33-40, 12th IEEE International Conference on Cloud Computing, CLOUD 2019, Milan, Italy, 07/8/19. https://doi.org/10.1109/CLOUD.2019.00018

Exploring potential for non-disruptive vertical auto scaling and resource estimation in kubernetes. / Rattihalli, Gourav; Govindaraju, Madhusudhan; Lu, Hui et al.
Proceedings - 2019 IEEE International Conference on Cloud Computing, CLOUD 2019 - Part of the 2019 IEEE World Congress on Services. ed. / Elisa Bertino; Carl K. Chang; Peter Chen; Ernesto Damiani; Michael Goul; Katsunori Oyama. IEEE Computer Society, 2019. p. 33-40 8814504 (IEEE International Conference on Cloud Computing, CLOUD; Vol. 2019-July).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Exploring potential for non-disruptive vertical auto scaling and resource estimation in kubernetes

AU - Rattihalli, Gourav

AU - Govindaraju, Madhusudhan

AU - Lu, Hui

AU - Tiwari, Devesh

PY - 2019/7

Y1 - 2019/7

N2 - Cloud platforms typically require users to provide resource requirements for applications so that resource managers can schedule containers with adequate allocations. However, the requirements for container resources often depend on numerous factors such as application input parameters, optimization flags, input files, and attributes that are specified for each run. So, it is complex for users to estimate the resource requirements for a given container accurately, leading to resource over-estimation that negatively affects overall utilization. We have designed a Resource Utilization Based Autoscaling System (RUBAS) that can dynamically adjust the allocation of containers running in a Kubernetes cluster. RUBAS improves upon the Kubernetes Vertical Pod Autoscaler (VPA) system non-disruptively by incorporating container migration. Our experiments use multiple scientific benchmarks. We analyze the allocation pattern of RUBAS with Kubernetes VPA. We compare the performance of container migration for in-place and remote node migration and we evaluate the overhead in RUBAS. Our results show that compared to Kubernetes VPA, RUBAS improves the CPU and memory utilization of the cluster by 10% and reduces the runtime by 15% with an overhead for each application ranging from 5% to 20%.

AB - Cloud platforms typically require users to provide resource requirements for applications so that resource managers can schedule containers with adequate allocations. However, the requirements for container resources often depend on numerous factors such as application input parameters, optimization flags, input files, and attributes that are specified for each run. So, it is complex for users to estimate the resource requirements for a given container accurately, leading to resource over-estimation that negatively affects overall utilization. We have designed a Resource Utilization Based Autoscaling System (RUBAS) that can dynamically adjust the allocation of containers running in a Kubernetes cluster. RUBAS improves upon the Kubernetes Vertical Pod Autoscaler (VPA) system non-disruptively by incorporating container migration. Our experiments use multiple scientific benchmarks. We analyze the allocation pattern of RUBAS with Kubernetes VPA. We compare the performance of container migration for in-place and remote node migration and we evaluate the overhead in RUBAS. Our results show that compared to Kubernetes VPA, RUBAS improves the CPU and memory utilization of the cluster by 10% and reduces the runtime by 15% with an overhead for each application ranging from 5% to 20%.

KW - Container migration

KW - Docker

KW - Docker Swarm

KW - Kubernetes

KW - VPA

KW - Vertical Scaling

UR - https://www.scopus.com/pages/publications/85072324127

U2 - 10.1109/CLOUD.2019.00018

DO - 10.1109/CLOUD.2019.00018

M3 - Conference contribution

T3 - IEEE International Conference on Cloud Computing, CLOUD

SP - 33

EP - 40

BT - Proceedings - 2019 IEEE International Conference on Cloud Computing, CLOUD 2019 - Part of the 2019 IEEE World Congress on Services

A2 - Bertino, Elisa

A2 - Chang, Carl K.

A2 - Chen, Peter

A2 - Damiani, Ernesto

A2 - Goul, Michael

A2 - Oyama, Katsunori

PB - IEEE Computer Society

T2 - 12th IEEE International Conference on Cloud Computing, CLOUD 2019

Y2 - 8 July 2019 through 13 July 2019

ER -

Rattihalli G, Govindaraju M, Lu H, Tiwari D. Exploring potential for non-disruptive vertical auto scaling and resource estimation in kubernetes. In Bertino E, Chang CK, Chen P, Damiani E, Goul M, Oyama K, editors, Proceedings - 2019 IEEE International Conference on Cloud Computing, CLOUD 2019 - Part of the 2019 IEEE World Congress on Services. IEEE Computer Society. 2019. p. 33-40. 8814504. (IEEE International Conference on Cloud Computing, CLOUD). doi: 10.1109/CLOUD.2019.00018

Exploring potential for non-disruptive vertical auto scaling and resource estimation in kubernetes

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this