Several models[19 21] for RAID have been proposed.[19] proposed STAMP (Storage Modeling for Power) by mapping the front-end workload to back-end disk activities, and the power consumption of the back-end activities is computed.
Given a specific workload w, we map the number of nodes n allocated to the Reader Tier to the performance RT by iterating over the allocation space and collecting corresponding performance measurements under each allocation candidate.
They defined an intelligent workload allocation system that efficiently mapped workloads to the best fitted resources in heterogeneous platforms.
For a single workload, simply setting and for the workload maps it into the stochastic framework and all the above proofs apply.
In summary, these steps represent the mapping, allocation, and execution of the workload based on the resources selected in the provisioning phase.
Further, the workload under evaluation is mapped into the hardware model to carry out the performance and energy consumption estimation.
In the workload, 20 Map tasks corresponding to the 1GB input size are equally distributed over 10 Slave nodes.
The whole approach only needs to define a group of meta-models and mapping rules, thus greatly reduces the workload of hand coding.
Mapping: Correspondence between resources required by the workload and resources provided by the cloud infrastructure.
To address this issue, we present a Partitioner, SP-Partitioner, that sits between the map and reduce stages to re-balance the workload of the reducers.
