An analytics-based envelope neural net approach to forecasting memory leakage in an enterprise applications environment

Nitin Khosla; D Sharma

doi:10.1007/978-3-030-42363-6_8

An analytics-based envelope neural net approach to forecasting memory leakage in an enterprise applications environment

Nitin Khosla, D Sharma

Information Systems

Research output: A Conference proceeding or a Chapter in Book › Conference contribution › peer-review

Abstract

This paper investigates a model to forecast memory-leak behavior in a big applications environment with short time frames, using neural networks. This process predicts the likelihood of a memory leakage in an IT infrastructure, with large number of enterprise applications running in a real world, while analyzing the work-load demand patterns hidden in the transactional data of the enterprise applications. This approach generates synthetic workload profiles and then based upon extracted insights, predict memory leakage behavior under peak load and stress conditions. The approach is investigated also for capacity tuning of the existing infrastructure and future planning of the required infrastructure. Co-learning, is used to train models and their use in analyzing the workload patterns. The workload prediction has enormous potential to optimize resources with minimal risk.

Original language	English
Title of host publication	International Conference on Innovation in Modern Science and Technology (ICIMSAT 2019): Intelligent Techniques and Applications in Science and Technology
Subtitle of host publication	Learning and Analytics in Intelligent Systems
Editors	Subhojit Dawn, Valentina Balas, Anna Esposito, Sadhan Gope
Place of Publication	Switzerland
Publisher	Springer
Pages	66-74
Number of pages	9
Volume	12
ISBN (Electronic)	9783030423636
ISBN (Print)	9783030423629
DOIs	https://doi.org/10.1007/978-3-030-42363-6_8
Publication status	Published - 3 Mar 2020
Event	International Conference on Innovation in Modern Science and Technology - , India Duration: 20 Sept 2019 → 21 Sept 2019 https://www.icimsat2019.com/

Publication series

Name	Intelligent Techniques and Applications in Science and Technology
Publisher	Springer
Volume	12
ISSN (Print)	2662-3447
ISSN (Electronic)	2662-3455

Conference

Conference	International Conference on Innovation in Modern Science and Technology
Abbreviated title	ICIMSAT 2019
Country/Territory	India
Period	20/09/19 → 21/09/19
Internet address	https://www.icimsat2019.com/

Access to Document

10.1007/978-3-030-42363-6_8

https://link.springer.com/content/pdf/10.1007%2F978-3-030-42363-6.pdf#page=82

Cite this

Khosla, N., & Sharma, D. (2020). An analytics-based envelope neural net approach to forecasting memory leakage in an enterprise applications environment. In S. Dawn, V. Balas, A. Esposito, & S. Gope (Eds.), International Conference on Innovation in Modern Science and Technology (ICIMSAT 2019): Intelligent Techniques and Applications in Science and Technology: Learning and Analytics in Intelligent Systems (Vol. 12, pp. 66-74). (Intelligent Techniques and Applications in Science and Technology; Vol. 12). Springer. https://doi.org/10.1007/978-3-030-42363-6_8

Khosla, Nitin ; Sharma, D. / An analytics-based envelope neural net approach to forecasting memory leakage in an enterprise applications environment. International Conference on Innovation in Modern Science and Technology (ICIMSAT 2019): Intelligent Techniques and Applications in Science and Technology: Learning and Analytics in Intelligent Systems. editor / Subhojit Dawn ; Valentina Balas ; Anna Esposito ; Sadhan Gope. Vol. 12 Switzerland : Springer, 2020. pp. 66-74 (Intelligent Techniques and Applications in Science and Technology).

@inproceedings{0a835f46d84a46d6a7ed3d3854b5a498,

title = "An analytics-based envelope neural net approach to forecasting memory leakage in an enterprise applications environment",

abstract = "This paper investigates a model to forecast memory-leak behavior in a big applications environment with short time frames, using neural networks. This process predicts the likelihood of a memory leakage in an IT infrastructure, with large number of enterprise applications running in a real world, while analyzing the work-load demand patterns hidden in the transactional data of the enterprise applications. This approach generates synthetic workload profiles and then based upon extracted insights, predict memory leakage behavior under peak load and stress conditions. The approach is investigated also for capacity tuning of the existing infrastructure and future planning of the required infrastructure. Co-learning, is used to train models and their use in analyzing the workload patterns. The workload prediction has enormous potential to optimize resources with minimal risk.",

keywords = "Performance engineering, Neural- networks, Semi-supervised learning, Load and stress testing, Co-learning, Machine learning",

author = "Nitin Khosla and D Sharma",

year = "2020",

month = mar,

day = "3",

doi = "10.1007/978-3-030-42363-6_8",

language = "English",

isbn = "9783030423629",

volume = "12",

series = "Intelligent Techniques and Applications in Science and Technology",

publisher = "Springer",

pages = "66--74",

editor = "Subhojit Dawn and Valentina Balas and Anna Esposito and Sadhan Gope",

booktitle = "International Conference on Innovation in Modern Science and Technology (ICIMSAT 2019): Intelligent Techniques and Applications in Science and Technology",

address = "Netherlands",

note = "International Conference on Innovation in Modern Science and Technology, ICIMSAT 2019 ; Conference date: 20-09-2019 Through 21-09-2019",

url = "https://www.icimsat2019.com/",

}

Khosla, N & Sharma, D 2020, An analytics-based envelope neural net approach to forecasting memory leakage in an enterprise applications environment. in S Dawn, V Balas, A Esposito & S Gope (eds), International Conference on Innovation in Modern Science and Technology (ICIMSAT 2019): Intelligent Techniques and Applications in Science and Technology: Learning and Analytics in Intelligent Systems. vol. 12, Intelligent Techniques and Applications in Science and Technology, vol. 12, Springer, Switzerland, pp. 66-74, International Conference on Innovation in Modern Science and Technology, India, 20/09/19. https://doi.org/10.1007/978-3-030-42363-6_8

An analytics-based envelope neural net approach to forecasting memory leakage in an enterprise applications environment. / Khosla, Nitin; Sharma, D.
International Conference on Innovation in Modern Science and Technology (ICIMSAT 2019): Intelligent Techniques and Applications in Science and Technology: Learning and Analytics in Intelligent Systems. ed. / Subhojit Dawn; Valentina Balas; Anna Esposito; Sadhan Gope. Vol. 12 Switzerland: Springer, 2020. p. 66-74 (Intelligent Techniques and Applications in Science and Technology; Vol. 12).

Research output: A Conference proceeding or a Chapter in Book › Conference contribution › peer-review

TY - GEN

T1 - An analytics-based envelope neural net approach to forecasting memory leakage in an enterprise applications environment

AU - Khosla, Nitin

AU - Sharma, D

PY - 2020/3/3

Y1 - 2020/3/3

N2 - This paper investigates a model to forecast memory-leak behavior in a big applications environment with short time frames, using neural networks. This process predicts the likelihood of a memory leakage in an IT infrastructure, with large number of enterprise applications running in a real world, while analyzing the work-load demand patterns hidden in the transactional data of the enterprise applications. This approach generates synthetic workload profiles and then based upon extracted insights, predict memory leakage behavior under peak load and stress conditions. The approach is investigated also for capacity tuning of the existing infrastructure and future planning of the required infrastructure. Co-learning, is used to train models and their use in analyzing the workload patterns. The workload prediction has enormous potential to optimize resources with minimal risk.

AB - This paper investigates a model to forecast memory-leak behavior in a big applications environment with short time frames, using neural networks. This process predicts the likelihood of a memory leakage in an IT infrastructure, with large number of enterprise applications running in a real world, while analyzing the work-load demand patterns hidden in the transactional data of the enterprise applications. This approach generates synthetic workload profiles and then based upon extracted insights, predict memory leakage behavior under peak load and stress conditions. The approach is investigated also for capacity tuning of the existing infrastructure and future planning of the required infrastructure. Co-learning, is used to train models and their use in analyzing the workload patterns. The workload prediction has enormous potential to optimize resources with minimal risk.

KW - Performance engineering, Neural- networks, Semi-supervised learning, Load and stress testing, Co-learning, Machine learning

U2 - 10.1007/978-3-030-42363-6_8

DO - 10.1007/978-3-030-42363-6_8

M3 - Conference contribution

SN - 9783030423629

VL - 12

T3 - Intelligent Techniques and Applications in Science and Technology

SP - 66

EP - 74

BT - International Conference on Innovation in Modern Science and Technology (ICIMSAT 2019): Intelligent Techniques and Applications in Science and Technology

A2 - Dawn, Subhojit

A2 - Balas, Valentina

A2 - Esposito, Anna

A2 - Gope, Sadhan

PB - Springer

CY - Switzerland

T2 - International Conference on Innovation in Modern Science and Technology

Y2 - 20 September 2019 through 21 September 2019

ER -

Khosla N, Sharma D. An analytics-based envelope neural net approach to forecasting memory leakage in an enterprise applications environment. In Dawn S, Balas V, Esposito A, Gope S, editors, International Conference on Innovation in Modern Science and Technology (ICIMSAT 2019): Intelligent Techniques and Applications in Science and Technology: Learning and Analytics in Intelligent Systems. Vol. 12. Switzerland: Springer. 2020. p. 66-74. (Intelligent Techniques and Applications in Science and Technology). doi: 10.1007/978-3-030-42363-6_8