Monitoring the Performance of Cloud Workload Through JConsole
The advancement of cloud computing has led to several challenges for researchers and developers. Monitoring the performance of the cloud workload is essential in order to evaluate complex applications deployed on the cloud. Traditional and automated sched
- PDF / 384,533 Bytes
- 10 Pages / 439.37 x 666.142 pts Page_size
- 5 Downloads / 220 Views
Abstract The advancement of cloud computing has led to several challenges for researchers and developers. Monitoring the performance of the cloud workload is essential in order to evaluate complex applications deployed on the cloud. Traditional and automated schedulers help in workload management. Web applications deployed on the cloud are among the categories of workload. Console-based applications provide mechanisms to handle key data processing tasks. This paper makes an attempt to deploy a Web application on the application server and study its performance using the Java performance monitoring tool, JConsole. Keywords Cloud computing
Workload Scheduling Monitoring
1 Introduction Cloud computing is the recent evolution we have seen to build and deliver services. Monitoring the performance of the cloud workload is required in order to evaluate complex multitier applications that are deployed on the cloud [1]. The applications deployed over the cloud behave differently and have their own characteristics. The challenges to the researchers and the developers are to provide workload specification. The applications in a cloud environment give rise to unpredicted workloads. The unpredicted workloads have to be attended by expanding the infrastructure. Expanding the infrastructure may not be a good solution as it would require a lot of investment. The virtualisation and dynamic scaling properties of the cloud help in providing a solution. The popular Java platform tools are often used for monitoring. The Java platform tools also provide information about performance and resource consumption [2]. Efficient management by analysing the performance and resource M.B. Vibha (&) Department of MCA, Dayananda Sagar College of Engineering, Bangalore, India e-mail: [email protected] M.B. Vibha R.R. Gondkar Bharathiar University, Coimbatore, India © Springer Science+Business Media Singapore 2016 N.R. Shetty et al. (eds.), Emerging Research in Computing, Information, Communication and Applications, DOI 10.1007/978-981-10-0287-8_50
533
534
M.B. Vibha and R.R. Gondkar
consumption will give an insight about the workload. The loader would be loaded with applications which are workloads and when a threshold is reached the performance of the application server indicates to expand the infrastructure. The workload monitoring can be either by a traditional approach or through automated schedulers. The traditional approaches are application specific, whereas automated schedulers help in managing unpredicted events. Schedulers can perform automatic submission of executions. The interfaces are used to define workflows and job dependencies. The priorities/queues control the execution order of unrelated jobs [3]. The additional features of schedulers include real-time scheduling based on external, unpredictable events, automatic restart and recovery of failures, and notifying operations personnel [3]. Furthermore, this paper discusses cloud workload and why JConsole is chosen for monitoring in Sects. 2 and 3, respectively. In the nex
Data Loading...
