Statistics-driven workload modeling for the Cloud
Authors:
- Archana GanapathiAGArchana Ganapathi
- Yanpei ChenYCYanpei Chen
- Armando FoxAFArmando Fox
- Randy KatzRKRandy Katz
- David PattersonDPDavid Patterson
- 1 January 2010
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 87-92
Abstract
A recent trend for data-intensive computations is to use pay-as-you-go execution environments that scale transparently to the user. However, providers of such environments must tackle the challenge of configuring their system to provide maximal performance while minimizing the cost of resources used. In this paper, we use statistical models to predict resource requirements for Cloud computing applications. Such a prediction framework can guide system design and deployment decisions such as scale, scheduling, and capacity. In addition, we present initial design of a workload generator that can be used to evaluate alternative configurations without the overhead of reproducing a real workload. This paper focuses on statistical modeling and its application to data-intensive workloads.Keywords
This publication has 4 references indexed in Scilit:
- Estimating the progress of MapReduce pipelines2010
- Pig latin2008
- MapReduceCommunications of the ACM, 2008
- The Google file systemACM SIGOPS Operating Systems Review, 2003