CPs have their system resources, such as a CPU core, memory, storage, and network bandwidth.
In this study, we mainly consider four resource types: CPU, memory, storage, and network bandwidth.
Each host server has different hardware resources (e.g., CPU, memory, storage) and network propagation delay values (10 5000 ms).
It performs the provisioning of resources (e.g. CPU time, memory, storage, and network bandwidth) to a vCell, interacts with the underlying layer, and performs additional global scheduling.
Of course, any live migration procedure must ensure consistency in the VM memory, storage, and network states before and after the transfer to the new hosting server.
The principles are similar to shared hosting, but hardware is virtualized and managed by a hypervisor that is able to administer servers as well as create partitions of CPU, memory, storage and network.
For example, memory and network I/O should be considered during VM placement as they have a significant effect in memory and network intensive applications.
CPU, Memory and network performances for a High Performance Computing Cluster (HPCC) is considered in [13].
Several metrics such as CPU, memory and network bandwidth consumption are captured using the sysstat tool[27].
Many of the world's leading supercomputer architectures are a hybrid of shared memory and network-distributed memory.
