A Survey and Taxonomy of Self-Aware and Self-Adaptive Cloud Autoscaling Systems

Chen, Tao; Bahsoon, Rami; Yao, Xin

doi:10.1145/3190507

Cited by 84 publications

(54 citation statements)

References 123 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, it is not difficult to build the model as classification, where the output is a discrete value, e.g., for security levels. In this work, we focus on the performance indicators that demands regression as they are arguably the most popular ones in existing work [22] [23].…”

Section: Background Related Work and Motivationmentioning

confidence: 99%

All Versus One: An Empirical Comparison on Retrained and Incremental Machine Learning for Modeling Performance of Adaptable Software

Chen

2019

2019 IEEE/ACM 14th International Symposium on Software Engineering for Adaptive and Self-Managing Systems (SEAMS)

Self Cite

View full text Add to dashboard Cite

Given the ever-increasing complexity of adaptable software systems and their commonly hidden internal information (e.g., software runs in the public cloud), machine learning based performance modeling has gained momentum for evaluating, understanding and predicting software performance, which facilitates better informed self-adaptations. As performance data accumulates during the run of the software, updating the performance models becomes necessary. To this end, there are two conventional modeling methods: the retrained modeling that always discard the old model and retrain a new one using all available data; or the incremental modeling that retains the existing model and tunes it using one newly arrival data sample. Generally, literature on machine learning based performance modeling for adaptable software chooses either of those methods according to a general belief, but they provide insufficient evidences or references to justify their choice. This paper is the first to report on a comprehensive empirical study that examines both modeling methods under distinct domains of adaptable software, 5 performance indicators, 8 learning algorithms and settings, covering a total of 1,360 different conditions. Our findings challenge the general belief, which is shown to be only partially correct, and reveal some of the important, statistically significant factors that are often overlooked in existing work, providing evidence-based insights on the choice.

show abstract

Section: Background Related Work and Motivationmentioning

confidence: 99%

All Versus One: An Empirical Comparison on Retrained and Incremental Machine Learning for Modeling Performance of Adaptable Software

Chen

2019

2019 IEEE/ACM 14th International Symposium on Software Engineering for Adaptive and Self-Managing Systems (SEAMS)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our goal is to provide a more complete solution that considers vertical scale elasticity for cloud resources management. 8…”

Section: Resultsmentioning

confidence: 99%

“…This new approach enables seeing the elasticity controller as an intrinsic entity of the cloud system. Therefore, monitoring tasks over the controlled cloud system enables considering it as "self-aware"; and the adaptation actions that are triggered in function of its state enables considering it as "self-adaptive" [8]. In addition, we propose a way to combine the different designed strategies to provide multiple cross-layer elastic patterns.…”

Section: Fig 3 Evaluation Of Cross-layer Elasticity Strategiesmentioning

confidence: 99%

Modeling and Evaluating Cross-layer Elasticity Strategies in Cloud Systems

Khebbeb

Hameurlain

Belala

2018

Model and Data Engineering

View full text Add to dashboard Cite

Clouds are complex systems that provide computing resources in an elastic way. Elasticity property allows their adaptation to input workload by (de)provisioning resources as the demand rises and drops. However, due to the numerous overlapping factors that impact their elasticity and the unpredictable nature of the workload, providing accurate action plans to manage cloud systems' elastic adaptations is a particularly challenging task. In this paper, we propose an approach based on Bigraphical Reactive Systems (BRS) to model cloud structures and their elastic behavior. We design elasticity strategies that operate at service and infrastructure cloud levels to manage the elastic adaptations. Besides, we provide a Maude encoding to permit generic executability and formal verification of the elastic behaviors. One step ahead, we show how the strategies can be combined at both levels to provide different high-level elastic behaviors. Finally, we evaluate the different cross-layer combinations using Queuing Theory.

show abstract

“…Related Surveys. Earlier surveys have been conducted by the authors [49,88,85,54,28,107,23,109], but the regular updates in cloud infrastructure and persistent research yielding the new research areas. There is a need to explore the present challenges and new research area in this field.…”

Section: Contribution Of the Studymentioning

confidence: 99%

Research on Auto-Scaling of Web Applications in Cloud: Survey, Trends and Future Directions

Singh

Gupta

Jyoti

et al. 2019

SCPE

View full text Add to dashboard Cite

Cloud computing emerging environment attracts many applications providers to deploy web applications on cloud data centers. The primary area of attraction is elasticity, which allows to auto-scale the resources on-demand. However, web applications usually have dynamic workload and hard to predict. Cloud service providers and researchers are working to reduce the cost while maintaining the Quality of Service (QoS). One of the key challenges for web application in cloud computing is auto-scaling. The auto-scaling in cloud computing is still in infancy and required detail investigation of taxonomy, approach and types of resources mapped to the current research. In this article, we presented the literature survey for auto-scaling techniques of web applications in cloud computing. This survey supports the research community to find the requirements in auto-scaling techniques. We present a taxonomy of reviewed articles with parameters such as auto-scaling techniques, approach, resources, monitoring tool, experiment, workload, and metric, etc. Based on the analysis, we proposed the new areas of research in this direction.

show abstract

A Survey and Taxonomy of Self-Aware and Self-Adaptive Cloud Autoscaling Systems

Cited by 84 publications

References 123 publications

All Versus One: An Empirical Comparison on Retrained and Incremental Machine Learning for Modeling Performance of Adaptable Software

All Versus One: An Empirical Comparison on Retrained and Incremental Machine Learning for Modeling Performance of Adaptable Software

Modeling and Evaluating Cross-layer Elasticity Strategies in Cloud Systems

Research on Auto-Scaling of Web Applications in Cloud: Survey, Trends and Future Directions

Contact Info

Product

Resources

About