A Modified MapReduce Framework for Cloud Computing
Lingying Zeng, Hao Wen Lin · 2012
Due to the heterogeneous, opaqueness and dynamic nature of Cloud Computing, existing MapReduce approach is not suitable for perform Parallel Computing on the cloud. In this paper, we propose a modified MapReduce framework which extracts the physical network topology information from the Virtual Machine Monitor (VMM) feature of VMs, in order to exploit dynamic resource allocations, and hence enable effective Parallel Computing within the cloud environment.