Using Performance Forecasting to Accelerate Elasticity

Paulo Moura; Fabio Kon; Spyros Voulgaris; Martinus Richardus van Steen

doi:10.1007/978-3-319-28448-4_2

Using Performance Forecasting to Accelerate Elasticity

Paulo Moura, Fabio Kon, Spyros Voulgaris, Martinus Richardus van Steen

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

1 Citation (Scopus)

202 Downloads (Pure)

Abstract

Cloud computing facilitates dynamic resource provisioning. The automation of resource management, known as elasticity, has been subject to much research. In this context, monitoring of a running service plays a crucial role, and adjustments are made when certain thresholds are crossed. On such occasions, it is common practice to simply add or remove resources. In this paper we investigate how we can predict the performance of a service to dynamically adjust allocated resources based on predictions. In other words, instead of “repairing‿ because a threshold has been crossed, we attempt to stay ahead and allocate an optimized amount of resources in advance. To do so, we need to have accurate predictive models that are based on workloads. We present our approach, based on the Universal Scalability Law, and discuss initial experiments.

Original language	Undefined
Title of host publication	Proceedings of the Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing (ARMS-CC 2015), Revised Selected Papers
Place of Publication	Berlin
Publisher	Springer
Pages	17-31
Number of pages	15
ISBN (Print)	978-3-319-28447-7
DOIs	https://doi.org/10.1007/978-3-319-28448-4_2
Publication status	Published - Jul 2015
Event	Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing, ARMS-CC 2015 - Donostia, Spain Duration: 20 Jul 2015 → 20 Jul 2015

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer Verlag
Volume	9438
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Workshop

Workshop	Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing, ARMS-CC 2015
Period	20/07/15 → 20/07/15
Other	20 July 2015

Keywords

EWI-26880
scalabilitymodeling
Elasticity
METIS-316852
IR-100082
Cloud computing
Performance Prediction

Access to Document

10.1007/978-3-319-28448-4_2

http://eprints.eemcs.utwente.nl/secure2/26880/02/chp%253A10.1007%252F978-3-319-28448-4_2.pdf

Cite this

Moura, P., Kon, F., Voulgaris, S., & van Steen, M. R. (2015). Using Performance Forecasting to Accelerate Elasticity. In Proceedings of the Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing (ARMS-CC 2015), Revised Selected Papers (pp. 17-31). (Lecture Notes in Computer Science; Vol. 9438). Springer. https://doi.org/10.1007/978-3-319-28448-4_2

@inproceedings{0cc891b976184481af6e92b01baea960,

title = "Using Performance Forecasting to Accelerate Elasticity",

abstract = "Cloud computing facilitates dynamic resource provisioning. The automation of resource management, known as elasticity, has been subject to much research. In this context, monitoring of a running service plays a crucial role, and adjustments are made when certain thresholds are crossed. On such occasions, it is common practice to simply add or remove resources. In this paper we investigate how we can predict the performance of a service to dynamically adjust allocated resources based on predictions. In other words, instead of “repairing‿ because a threshold has been crossed, we attempt to stay ahead and allocate an optimized amount of resources in advance. To do so, we need to have accurate predictive models that are based on workloads. We present our approach, based on the Universal Scalability Law, and discuss initial experiments.",

keywords = "EWI-26880, scalabilitymodeling, Elasticity, METIS-316852, IR-100082, Cloud computing, Performance Prediction",

author = "Paulo Moura and Fabio Kon and Spyros Voulgaris and {van Steen}, {Martinus Richardus}",

note = "10.1007/978-3-319-28448-4_2 ; Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing, ARMS-CC 2015 ; Conference date: 20-07-2015 Through 20-07-2015",

year = "2015",

month = jul,

doi = "10.1007/978-3-319-28448-4_2",

language = "Undefined",

isbn = "978-3-319-28447-7",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "17--31",

booktitle = "Proceedings of the Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing (ARMS-CC 2015), Revised Selected Papers",

address = "Germany",

}

Moura, P, Kon, F, Voulgaris, S & van Steen, MR 2015, Using Performance Forecasting to Accelerate Elasticity. in Proceedings of the Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing (ARMS-CC 2015), Revised Selected Papers. Lecture Notes in Computer Science, vol. 9438, Springer, Berlin, pp. 17-31, Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing, ARMS-CC 2015, 20/07/15. https://doi.org/10.1007/978-3-319-28448-4_2

Using Performance Forecasting to Accelerate Elasticity. / Moura, Paulo; Kon, Fabio; Voulgaris, Spyros et al.
Proceedings of the Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing (ARMS-CC 2015), Revised Selected Papers. Berlin: Springer, 2015. p. 17-31 (Lecture Notes in Computer Science; Vol. 9438).

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

TY - GEN

T1 - Using Performance Forecasting to Accelerate Elasticity

AU - Moura, Paulo

AU - Kon, Fabio

AU - Voulgaris, Spyros

AU - van Steen, Martinus Richardus

N1 - 10.1007/978-3-319-28448-4_2

PY - 2015/7

Y1 - 2015/7

N2 - Cloud computing facilitates dynamic resource provisioning. The automation of resource management, known as elasticity, has been subject to much research. In this context, monitoring of a running service plays a crucial role, and adjustments are made when certain thresholds are crossed. On such occasions, it is common practice to simply add or remove resources. In this paper we investigate how we can predict the performance of a service to dynamically adjust allocated resources based on predictions. In other words, instead of “repairing‿ because a threshold has been crossed, we attempt to stay ahead and allocate an optimized amount of resources in advance. To do so, we need to have accurate predictive models that are based on workloads. We present our approach, based on the Universal Scalability Law, and discuss initial experiments.

AB - Cloud computing facilitates dynamic resource provisioning. The automation of resource management, known as elasticity, has been subject to much research. In this context, monitoring of a running service plays a crucial role, and adjustments are made when certain thresholds are crossed. On such occasions, it is common practice to simply add or remove resources. In this paper we investigate how we can predict the performance of a service to dynamically adjust allocated resources based on predictions. In other words, instead of “repairing‿ because a threshold has been crossed, we attempt to stay ahead and allocate an optimized amount of resources in advance. To do so, we need to have accurate predictive models that are based on workloads. We present our approach, based on the Universal Scalability Law, and discuss initial experiments.

KW - EWI-26880

KW - scalabilitymodeling

KW - Elasticity

KW - METIS-316852

KW - IR-100082

KW - Cloud computing

KW - Performance Prediction

U2 - 10.1007/978-3-319-28448-4_2

DO - 10.1007/978-3-319-28448-4_2

M3 - Conference contribution

SN - 978-3-319-28447-7

T3 - Lecture Notes in Computer Science

SP - 17

EP - 31

BT - Proceedings of the Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing (ARMS-CC 2015), Revised Selected Papers

PB - Springer

CY - Berlin

T2 - Second International Workshop on Adaptive Resource Management and Scheduling for Cloud Computing, ARMS-CC 2015

Y2 - 20 July 2015 through 20 July 2015

ER -