Quantitative system performance: computer system analysis using queueing network models
Quantitative system performance: computer system analysis using queueing network models
JMT: performance engineering tools for system modeling
ACM SIGMETRICS Performance Evaluation Review
Linux Journal
The Performance Cockpit Approach: A Framework For Systematic Performance Evaluations
SEAA '10 Proceedings of the 2010 36th EUROMICRO Conference on Software Engineering and Advanced Applications
Flexible CPU Provisioning in Clouds: A New Source of Performance Unpredictability
QEST '12 Proceedings of the 2012 Ninth International Conference on Quantitative Evaluation of Systems
Hi-index | 0.00 |
Cloud computing is an ever growing reality nowadays. The possibility of leasing and provisioning resources is an important alternative to the construction of dedicated data centers. At the same time, multi-core and many-core architectures are becoming more and more popular and widely available. However, such architectures present several challenges due to their high complexity. In particular it is essential to be able to study, test, model and simulate various aspects of these new environments, with suitable tools. In this paper we present jBM, a simple tool for the benchmarking the CPU of new architectures like multi-core or cloud systems. The tool consists of a multi-class workload generator and service centers that can be distributed along different nodes. The applicability of the tool is proven through several applications to evaluate the performance of virtual machines on Amazon EC2.