JTangCMS: An efficient monitoring system for cloud platforms

详细信息    查看全文

• 作者：Xingjian Lu^a ; ^b ; ^{luxj@ecust.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Jianwei Yin^b ; ^{zjuyjw@zju.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Neal N. Xiong^c ; ^d ; ^{xiongnaixue@gmail.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Shuiguang Deng^b ; ^{dengsg@zju.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Gaoqi He^a ; ^{hegaoqi@ecust.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Huiqun Yu^a ; ^{yhq@ecust.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae
• 关键词：Cloud monitoring ; Data distribution service ; Complex event processing
• 刊名：Information Sciences
• 出版年：2016
• 出版时间：20 November 2016
• 年：2016
• 卷：370-371
• 期：Complete
• 页码：402-423
• 全文大小：2588 K

文摘

Cloud computing has attracted increasing interest in industry and academia. However, due to the constantly expanding scale of cloud platforms, the monitoring of clouds encounters a number of critical challenges related to flexibility, scalability, efficiency, and performance. In this paper, we present JTang Cloud Monitoring System (JTangCMS), an efficient monitoring system for cloud platforms. Our contributions cover the collection, delivery, and processing of monitoring data. For data collection, a flexible and scalable agent is implemented with pluggable monitoring components to collect runtime information from different entities. For data delivery, an efficient and robust data dissemination framework is implemented to transmit the runtime information reliably with high throughput and low latency, based on the Data Distribution Service (DDS). For data processing, a cloud action platform is developed to support cloud management decision-making, based on complex event processing (CEP). Finally, detailed experimental evaluations show the feasibility and efficiency of JTangCMS.