An Approach of Collecting Performance Anomaly Dataset for NFV Infrastructure

Du, Qingfeng; He, Yu Ting; Xie, Tianfa; Yin, Kanglin; Qiu, Juan

doi:10.1007/978-3-030-05057-3_5

Cited by 5 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [14], the authors evaluate several supervised learning approaches for Anomaly Detection (AD) by injecting faults in a Kubernetes cluster. Similarly, in [13], the authors also evaluate supervised learning techniques for off-line AD in an NFV environment.…”

Section: Related Workmentioning

confidence: 99%

“…Given the abundance of operational data, it is natural to seek for data-driven approaches like ML, that can augment the capabilities of operators to ''navigate'' the zillions of available time-series and logs. For instance, considering the AD problem, many works in the literature leverage on either supervised [13], [14] or unsupervised [15]- [17] ML algorithms to detect early symptoms of anomalous conditions at different levels of a cloud infrastructure. Similarly, many recent works [18]- [20] propose time-series forecasting techniques to anticipate the evolution of workloads and scale compute resources (e.g., VMs or containers) accordingly.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A 2-Phase Strategy for Intelligent Cloud Operations

Lanciano,

Andreoli,

Cucinotta

et al. 2023

IEEE Access

View full text Add to dashboard Cite

When operating large cloud computing infrastructures, ensuring healthiness of physical resources and software components is of paramount importance to meet the demanding service levels expected by customers. This is only possible using automations that can detect anomalies and alert the on-call personnel, or trigger healing procedures. In production-grade deployments, such automations are generally based on static thresholds or predefined pattern-matching rules, checked against relevant metrics and logs. Defining and maintaining them is cumbersome and, as the infrastructure grows, they need continuous adjustments. To tackle this problem, we propose an intelligent automation system for cloud operations that learns, from what operators have done in the past, what actions should be applied in response to the observed anomalies. Such system is designed to operate elastic groups of cloud instances realizing typical (replicated) cloud services. The mechanism is based on a 2-phase machine learning pipeline, composed of: a first, lighter, model that automatically detects anomalous patterns, based on past observations of the normal behavior, causing activation of the second, more involved, model; this is a model that recommends specific corrective actions, based on historical operational data reporting the actions applied to heal the faulty components. The approach was validated on an OpenStack deployment, where we deployed both a synthetic application and a multi-node Cassandra NoSQL data-store, and injected different types of anomalies while these systems were exercised using synthetic workloads. For both applications, we obtained a remarkable accuracy (mostly beyond 90%, and also going beyond 95% in some cases), for the anomaly detection and corrective action recommendation tasks, by applying the models on the respective test sets. This allows us to conclude that the presented mechanism constitutes an efficient and effective technique to help operating cloud services in presence of a number of faults, albeit the types and heterogeneity of faulty conditions might be expanded in future evolutions of the framework. The implementation and the material needed to reproduce our results are available under an open-source license.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A 2-Phase Strategy for Intelligent Cloud Operations

Lanciano,

Andreoli,

Cucinotta

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Anomaly detection has been widely studied in the context of NFV and cloud operations, often with the help of fault injection techniques. For example, in [18], researchers have built a data set injecting faults in a Kubernetes cluster, and evaluating different techniques for anomaly detection based on supervised machine learning (ML), including support vector machines (SVMs), nearest neighbor, naive Bayes and random forests. SVMs have also been used in [19] for on-line detection of anomalies in data from transmissions in Wireless Sensor Networks.…”

Section: Related Workmentioning

confidence: 99%

Near Real-Time Anomaly Detection in NFV Infrastructures

Derstepanians

Vannucci

Cucinotta

et al. 2022

2022 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)

View full text Add to dashboard Cite

This paper presents a scalable cloud-based architecture for near real-time anomaly detection in the Vodafone NFV infrastructure, spanning across multiple data centers in 11 European countries. Our solution aims at processing in real-time system-level data coming from the monitoring subsystem of the infrastructure, raising alerts to operators as soon as the incoming data presents anomalous patterns. A number of different anomaly detection techniques have been implemented for the proposed architecture, and results from their comparative evaluation are reported, based on real monitoring data coming from one of the monitored data centers, where a number of interesting anomalies have been manually identified. Part of this labelled data-set is also released under an open data license, for possible reuse by other researchers.

show abstract

Fault Diagnosis in the Network Function Virtualization: A Survey, Taxonomy, and Future Directions

Li,

Qi,

et al. 2024

IEEE Internet Things J.

View full text Add to dashboard Cite

An Approach of Collecting Performance Anomaly Dataset for NFV Infrastructure

Cited by 5 publications

References 11 publications

A 2-Phase Strategy for Intelligent Cloud Operations

A 2-Phase Strategy for Intelligent Cloud Operations

Near Real-Time Anomaly Detection in NFV Infrastructures

Fault Diagnosis in the Network Function Virtualization: A Survey, Taxonomy, and Future Directions

Contact Info

Product

Resources

About