Affiliation:
1. The University of Melbourne, VIC, Australia
Abstract
The dynamic nature of the cloud environment has made the distributed resource management process a challenge for cloud service providers. The importance of maintaining quality of service in accordance with customer expectations and the highly dynamic nature of cloud-hosted applications add new levels of complexity to the process. Advances in big-data learning approaches have shifted conventional static capacity planning solutions to complex performance-aware resource management methods. It is shown that the process of decision-making for resource adjustment is closely related to the behavior of the system, including the utilization of resources and application components. Therefore, a continuous monitoring of system attributes and performance metrics provides the raw data for the analysis of problems affecting the performance of the application. Data analytic methods, such as statistical and machine-learning approaches, offer the required concepts, models, and tools to dig into the data and find general rules, patterns, and characteristics that define the functionality of the system. Obtained knowledge from the data analysis process helps to determine the changes in the workloads, faulty components, or problems that can cause system performance to degrade. A timely reaction to performance degradation can avoid violations of service level agreements, including performing proper corrective actions such as auto-scaling or other resource adjustment solutions. In this article, we investigate the main requirements and limitations of cloud resource management, including a study of the approaches to workload and anomaly analysis in the context of performance management in the cloud. A taxonomy of the works on this problem is presented that identifies main approaches in existing research from the data analysis side to resource adjustment techniques. Finally, considering the observed gaps in the general direction of the reviewed works, a list of these gaps is proposed for future researchers to pursue.
Publisher
Association for Computing Machinery (ACM)
Subject
General Computer Science,Theoretical Computer Science
Cited by
12 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献