Workloads on clouds are set to be as efficient as possible. However, common problems with the operation of virtual systems arise over time. Individual virtual machines (VMs) that run the workloads are supposed to have the right levels of processing power, memory, and disk space. Sometimes, though, they do not use what has been allocated to them, which results in inefficiency.
In workload rightsizing, cloud administrators assess the virtual CPU processing power allocated to a workload. They also look at the assigned RAM and virtual disk space, and what the workload uses.
Although administrators can use manual workload rightsizing, there are software tools available that provide this kind of administration. These tools can either suggest manual pre-provisioning of resources or, in some cases, change the provisioning automatically.
Rightsizing is important for a VM. Here are some benefits:
The processes inside the Guest OS may experience less ping-pong. The Guest OS may not be aware of the NUMA nature of the physical motherboard, and think it has a uniform structure. It may move processes within its own CPUs, as it assumes it has no performance impact. If the vCPUs are spread into different NUMA node, example a 20 vCPU on a box with 2-socket and 20 cores, it can experience the ping-pong effect.
Lower risk of NUMA effect. Lower risk that the RAM or CPU is spread over a single socket. Due to NUMA architecture, the performance will not be as good.
Lower co-stop and ready time. Even if not all vCPU is used by the application, the Guest OS will still demand all the vCPU be provided by the hypervisor.
Faster snapshot time, especially if memory snapshot is included.
Faster boot time. If a VM does not have a reservation, vSphere will create a swap file the size of the configured RAM. This can impact the boot time if the storage subsystem is slow.
Faster vMotion. Windows and Linux use memory as cache. The more it has, the more it uses, all else being equal.
