Application of Cloud Distributed Computing on Autonomic Resource Allocation
Cloud computing is the latest trend in computing service where computing is leased out as a service on a pay-per-use basis. This has a huge potential in providing scientific computing to those users who do not have access to traditional resources like grids, clusters, etc. The authors study the cloud framework Eucalyptus and set it up for the purpose for scientific computing. The chosen problem matrix multiplication is set up using the message passing interface, a leading standard for message passing libraries. The experiments are done on and the resulting response times are observed. These response times are modeled as functions of the size of the jobs and the resources allocated.