A cloud administrator is provisioning five VMs, each with at least 8 GB of RAM, on a hypervisor with 32 GB of RAM. Which feature should be used to manage this constraint?

A cloud administrator is provisioning five VMs, each with at least 8 GB of RAM, on a hypervisor with 32 GB of RAM. Which feature should be used to manage this constraint?

• A. Memory overcommitment
• B. Thin-provisioned model
• C. Process scheduling
• D. Hyperthreading

Answer: (C)

Memory overcommitment is the technique used when the total RAM requested by all VMs exceeds the host’s physical memory. The hypervisor can allocate more virtual RAM to each VM than is physically available and reclaim unused memory as needed, using methods like ballooning, compression, or paging. In this scenario, five VMs each asking for 8 GB totals 40 GB, which exceeds the 32 GB on the host, so memory overcommitment is the way to manage the constraint. It allows VMs to start and run while the system reclaims memory under pressure, though it can impact performance if memory contention becomes high. The other options don’t address the RAM constraint: a thin-provisioned model is about storage capacity, not memory; process scheduling governs CPU time sharing among processes/VMs rather than memory allocation; hyperthreading is a CPU feature that doesn’t manage RAM.

Memory overcommitment is the technique used when the total RAM requested by all VMs exceeds the host’s physical memory. The hypervisor can allocate more virtual RAM to each VM than is physically available and reclaim unused memory as needed, using methods like ballooning, compression, or paging. In this scenario, five VMs each asking for 8 GB totals 40 GB, which exceeds the 32 GB on the host, so memory overcommitment is the way to manage the constraint. It allows VMs to start and run while the system reclaims memory under pressure, though it can impact performance if memory contention becomes high.

The other options don’t address the RAM constraint: a thin-provisioned model is about storage capacity, not memory; process scheduling governs CPU time sharing among processes/VMs rather than memory allocation; hyperthreading is a CPU feature that doesn’t manage RAM.