Affiliation:
1. University of Southern California
2. Microsoft Research, India
3. University of California, San Diego
Abstract
Modeling system performance and workload characteristics has become essential for efficiently provisioning Internet services and for accurately predicting future resource requirements on anticipated workloads. The accuracy of these models benefits substantially by differentiating among categories of requests based on their resource usage characteristics. However, categorizing requests and their resource demands often requires significantly more monitoring infrastructure. In this paper, we describe a method to automatically differentiate and categorize requests without requiring sophisticated monitoring techniques. Using machine learning, our method requires only aggregate measures such as total number of requests and the total CPU and network demands, and does not assume prior knowledge of request categories or their individual resource demands. We explore the feasibility of our method on the .Net PetShop 4.0 benchmark application, and show that it works well while being lightweight, generic, and easily deployable.
Publisher
Association for Computing Machinery (ACM)
Subject
Computer Networks and Communications,Hardware and Architecture,Software
Reference23 articles.
1. Capacity planning tools for web and grid environments
2. The FastICA package for Matlab and R. http://www.cis.hut.fi/projects/ica/fastica/. The FastICA package for Matlab and R. http://www.cis.hut.fi/projects/ica/fastica/.
Cited by
25 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
1. Estimation of Sharing Dependencies in Personal Storage Clouds Using Ensemble Learning Approaches;Operationalizing Multi-Cloud Environments;2021-09-18
2. SARDE;ACM Transactions on Autonomous and Adaptive Systems;2021-06
3. Learning Queuing Networks via Linear Optimization;Proceedings of the ACM/SPEC International Conference on Performance Engineering;2021-04-09
4. WITHDRAWN: A review on modeling techniques of quality-of-service;Materials Today: Proceedings;2021-02
5. Learning Queuing Networks by Recurrent Neural Networks;Proceedings of the ACM/SPEC International Conference on Performance Engineering;2020-04-20