Predicting Failures of Autoscaling Distributed Applications
Proceedings of the ACM on Software Engineering · 12.07.2024 · 1 (FSE) · 1960-1981
DOI: 10.1145/3660794 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the ACM on Software Engineering, online erschienen am 12.07.2024. Verfasst von Giovanni Denaro und 5 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: MUR, Ministero Università e Ricerca, SNF Swiss National Foundation.
Abstract (Originalsprache)
Predicting failures in production environments allows service providers to activate countermeasures that prevent harming the users of the applications. The most successful approaches predict failures from error states that the current approaches identify from anomalies in time series of fixed sets of KPI values collected at runtime. They cannot handle time series of KPI sets with size that varies over time. Thus these approaches work with applications that run on statically configured sets of components and computational nodes, and do not scale up to the many popular cloud applications that exploit autoscaling. This paper proposes P reface , a novel approach to predict failures in cloud applications that exploit autoscaling. P reface originally augments the neural-network-based failure predictors successfully exploited to predict failures in statically configured applications, with a R ectifier layer that handles KPI sets of highly variable size as the ones collected in cloud autoscaling applications, and reduces those KPIs to a set of rectified-KPIs of fixed size that can be fed to the neural-network predictor. The P reface R ectifier computes the rectified-KPIs as descriptive statistics of the original KPIs, for each logical component of the target application. The descriptive statistics shrink the highly variable sets of KPIs collected at different timestamps to a fixed set of values compatible with the input nodes of the neural-network failure predictor. The neural network can then reveal anomalies that correspond to error states, before they propagate to failures that harm the users of the applications. The experiments on both a commercial application and a widely used academic exemplar confirm that P reface can indeed predict many harmful failures early enough to activate proper countermeasures.
Förderung
- MUR, Ministero Università e Ricerca
- SNF Swiss National Foundation
Zitieren
APA
Denaro, G., Moussa, N. E., Heydarov, R., Lomio, F., Pezzè, M., & Qiu, K. (2024). Predicting Failures of Autoscaling Distributed Applications. Proceedings of the ACM on Software Engineering, 1(FSE), 1960-1981. https://doi.org/10.1145/3660794
Vancouver
Denaro G, Moussa NE, Heydarov R, Lomio F, Pezzè M, Qiu K. Predicting Failures of Autoscaling Distributed Applications. Proceedings of the ACM on Software Engineering. 2024;1(FSE):1960-1981. doi:10.1145/3660794
BibTeX
@article{Denaro2024,
author = {Denaro, Giovanni and Moussa, Noura El and Heydarov, Rahim and Lomio, Francesco and Pezzè, Mauro and Qiu, Ketai},
title = {Predicting Failures of Autoscaling Distributed Applications},
journal = {Proceedings of the ACM on Software Engineering},
year = {2024},
volume = {1},
number = {FSE},
pages = {1960--1981},
doi = {10.1145/3660794}
}
EUVIMED