%0 Conference Paper %1 2022vargisa %A Vargis, Tom Richard %A Ghiasvand, Siavash %B The International Conference for High Performance Computing, Networking, Storage, and Analysis %C Dallas, Texas, USA %D 2022 %K Parallel, myOwn from:ghiasvan Learning and Computer Cluster Distributed, Computing, Machine Science %T A Light-weight and Unsupervised Method for Near Real-time Behavioral Analysis using Operational Data Measurement %U https://sc22.supercomputing.org/proceedings/tech_poster/tech_poster_pages/rpost131.html %X Monitoring the status of large computing systems is essential to identify unexpected behavior and improve their performance and uptime. However, due to the large-scale and distributed design of such computing systems as well as a large number of monitoring parameters, automated monitoring methods should be applied. Such automatic monitoring methods should also have the ability to adapt themselves to the continuous changes in the computing system. In addition, they should be able to identify behavioral anomalies in useful time, to perform appropriate reactions. This work proposes a general lightweight and unsupervised method for near real-time anomaly detection using operational data measurement on large computing systems. The proposed model requires as little as 4 hours of data and 50 epochs for each training process to accurately resemble the behavioral pattern of computing systems.

@inproceedings{2022vargisa, abstract = {Monitoring the status of large computing systems is essential to identify unexpected behavior and improve their performance and uptime. However, due to the large-scale and distributed design of such computing systems as well as a large number of monitoring parameters, automated monitoring methods should be applied. Such automatic monitoring methods should also have the ability to adapt themselves to the continuous changes in the computing system. In addition, they should be able to identify behavioral anomalies in useful time, to perform appropriate reactions. This work proposes a general lightweight and unsupervised method for near real-time anomaly detection using operational data measurement on large computing systems. The proposed model requires as little as 4 hours of data and 50 epochs for each training process to accurately resemble the behavioral pattern of computing systems.}, added-at = {2024-12-10T16:31:24.000+0100}, address = {Dallas, Texas, USA}, author = {Vargis, Tom Richard and Ghiasvand, Siavash}, biburl = {https://puma.scadsai.uni-leipzig.de/bibtex/269af9710e324552260d22bac37dfbbf2/scads.ai}, booktitle = {The {International} {Conference} for {High} {Performance} {Computing}, {Networking}, {Storage}, and {Analysis}}, copyright = {All rights reserved}, file = {arXiv Fulltext PDF:D\:\\Documents\\Zotero\\storage\\BW4NRA2U\\Vargis and Ghiasvand - 2024 - A Light-weight and Unsupervised Method for Near Re.pdf:application/pdf;arXiv.org Snapshot:D\:\\Documents\\Zotero\\storage\\K6EWK7VR\\2402.html:text/html}, interhash = {853dca417d820d5b2702e4559c52db04}, intrahash = {69af9710e324552260d22bac37dfbbf2}, keywords = {Parallel, myOwn from:ghiasvan Learning and Computer Cluster Distributed, Computing, Machine Science}, month = jan, note = {arXiv:2402.05114 [cs]}, timestamp = {2024-12-10T16:31:24.000+0100}, title = {A {Light}-weight and {Unsupervised} {Method} for {Near} {Real}-time {Behavioral} {Analysis} using {Operational} {Data} {Measurement}}, url = {https://sc22.supercomputing.org/proceedings/tech_poster/tech_poster_pages/rpost131.html}, urldate = {2024-08-20}, year = 2022 }