Comparison of different neural network architectures for detecting anomalies of storage indicators

Vadim Moshkin; Dmitry Kurilo; Ilya Andreev; Yulia Gavrilova

doi:10.1117/12.3025021

20 February 2024 Comparison of different neural network architectures for detecting anomalies of storage indicators

Vadim Moshkin, Dmitry Kurilo, Ilya Andreev, Yulia Gavrilova

Proceedings Volume 13065, Third International Conference on Optics, Computer Applications, and Materials Science (CMSD-III 2023); 130650N (2024) https://doi.org/10.1117/12.3025021
Event: Third International Conference on Optics, Computer Applications, and Materials Science (CMSD-III 2023), 2023, Dushanbe, Tajikistan

Abstract

This article presents a comparison of the effectiveness of using different architectures of neural networks in solving the problem of detecting anomalies in time series. The performance characteristics of data storage systems as a subject area are presented. Data is taken from storage performance logs. The article compares the effectiveness of LSTM, SVM, Autoencoder, CNN, RNN architectures, and also presents the developed LSTM architecture. This neural network shows the highest results as a result of experiments. In the future, it is planned to refine the algorithm by interpreting the results using semantic knowledge bases based on ontologies and rules.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Vadim Moshkin, Dmitry Kurilo, Ilya Andreev, and Yulia Gavrilova "Comparison of different neural network architectures for detecting anomalies of storage indicators", Proc. SPIE 13065, Third International Conference on Optics, Computer Applications, and Materials Science (CMSD-III 2023), 130650N (20 February 2024); https://doi.org/10.1117/12.3025021

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available