Using a multi-agent system and artificial intelligence for monitoring and improving the cloud performance and security

Grzonka, Daniel; Jakóbik, Agnieszka; Kołodziej, Joanna; Pllana, Sabri

doi:10.1016/j.future.2017.05.046

Cited by 81 publications

(49 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Artificial neural network (ANN) [30,41,43,44], regression (LR, QR, PR) [9,24,36,52,54], support vector machines (SVM) [23,51,76], and decisiontrees (DT) [16, …”

Section: Rq2: Software Optimization Algorithms Used For Run-time Dynamentioning

confidence: 99%

Using meta-heuristics and machine learning for software optimization of parallel computing systems: a systematic literature review

et al. 2018

Self Cite

View full text Add to dashboard Cite

While modern parallel computing systems offer high performance, utilizing these powerful computing resources to the highest possible extent demands advanced knowledge of various hardware architectures and parallel programming models. Furthermore, optimized software execution on parallel computing systems demands consideration of many parameters at compile-time and run-time. Determining the optimal set of parameters in a given execution context is a complex task, and therefore to address this issue researchers have proposed different approaches that use heuristic search or machine learning. In this paper, we undertake a systematic literature review to aggregate, analyze and classify the existing software optimization methods for parallel computing systems. We review approaches that use machine learning or meta-heuristics for software optimization at compile-time and run-time. we discuss challenges and future research directions. The results of this study may help to better understand the state-of-the-art techniques that use machine learning and meta-heuristics to deal with the complexity of software optimization for parallel computing systems. Furthermore, it may aid in understanding the limitations of existing approaches and identification of areas for improvement.

show abstract

“…Artificial neural network (ANN) [30,41,43,44], regression (LR, QR, PR) [9,24,36,52,54], support vector machines (SVM) [23,51,76], and decisiontrees (DT) [16, …”

Section: Rq2: Software Optimization Algorithms Used For Run-time Dynamentioning

confidence: 99%

Using meta-heuristics and machine learning for software optimization of parallel computing systems: a systematic literature review

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…In our previous work (Cai et al 2017b), we have proposed to apply DAGGTAX for the analysis of DAP informally, while in this work, we propose to formally analyze DAGGTAX-based design with tool support. Recently, Grzonka et al (2017) have proposed a Self-Adaptive Resource Management System (SARMS) that achieves auto scaling based on CPU and memory usage. Their work focuses on the monitoring of resource usages as a whole, while our case study implements the monitoring and aggregation in several layers, including the virtual machines, the VNFC, and the applications.…”

Section: Related Workmentioning

confidence: 99%

Tool-supported design of data aggregation processes in cloud monitoring systems

Cai

Gallina

Nyström

et al. 2018

J Ambient Intell Human Comput

View full text Add to dashboard Cite

Efficient monitoring of a cloud system involves multiple aggregation processes and large amounts of data with various and interdependent requirements. A thorough understanding and analysis of the characteristics of data aggregation processes can help to improve the software quality and reduce development cost. In this paper, we propose a systematic approach for designing data aggregation processes in cloud monitoring systems. Our approach applies a feature-oriented taxonomy called DAGGTAX (Data AGGregation TAXonomy) to systematically specify the features of the designed system, and SAT-based analysis to check the consistency of the specifications. Following our approach, designers first specify the data aggregation processes by selecting and composing the features from DAGGTAX. These specified features, as well as design constraints, are then formalized as propositional formulas, whose consistency is checked by the Z3 SAT solver. To support our approach, we propose a design tool called SAFARE (SAt-based Feature-oriented dAta aggREgation design), which implements DAG-GTAX-based specification of data aggregation processes and design constraints, and integrates the state-of-the-art solver Z3 for automated analysis. We also propose a set of general design constraints, which are integrated by default in SAFARE. The effectiveness of our approach is demonstrated via a case study provided by industry, which aims to design a cloud monitoring system for video streaming. The case study shows that DAGGTAX and SAFARE can help designers to identify reusable features, eliminate infeasible design decisions, and derive crucial system parameters.

show abstract

“…Since its inception, distributed control of a multiagent network instantly became a popular topic in the community, witnessed by its broad applications in diverse areas, such as formation control, smart grid, and artificial intelligence, to name a few. As is well known, one common goal in these venues is to ensure that all networked systems reach an agreement/consensus with only local interactions in the presence of intermittent communications .…”

Section: Introductionmentioning

confidence: 99%

Dynamic average consensus with topology balancing under a directed graph

Xin

Wang

et al. 2019

Intl J Robust & Nonlinear

View full text Add to dashboard Cite

Summary In this paper, the distributed average tracking problem is studied on the premise of a strongly connected directed graph. To this end, we propose a weight balance strategy that could potentially make the adjacency matrix doubly stochastic for any strongly connected directed graph. The proposed scheme is fully distributive with finite time convergence and we again prove that network connectivity (described by the first left eigenvector) is instrumental in networked control systems. Then, a discrete‐time average tracking observer is introduced to ensure that all networked systems can track the average of the reference signals with bounded error. Simulation results verify the effectiveness of the proposed methods.

show abstract

Using a multi-agent system and artificial intelligence for monitoring and improving the cloud performance and security

Cited by 81 publications

References 22 publications

Using meta-heuristics and machine learning for software optimization of parallel computing systems: a systematic literature review

Using meta-heuristics and machine learning for software optimization of parallel computing systems: a systematic literature review

Tool-supported design of data aggregation processes in cloud monitoring systems

Dynamic average consensus with topology balancing under a directed graph

Contact Info

Product

Resources

About