NCP-MCI-6.10 · Question #121
An administrator is managing an environment based on two different AHV-based and ESXi-based clusters. Workloads are evenly distributed and in a healthy state. A Linux VM running on ESXi is not…
The correct answer is B. Migrate the VM to the AHV cluster. Migrating the VM to the AHV cluster is the easiest way to test whether the Nutanix native storage stack (ADSF/AHV) offers better storage performance than the ESXi-based configuration. Nutanix Move can perform cross-hypervisor migrations with minimal downtime. If performance…
Question
An administrator is managing an environment based on two different AHV-based and ESXi-based clusters. Workloads are evenly distributed and in a healthy state. A Linux VM running on ESXi is not performing well at the storage level and is configured as follows:
- VCPU: 8
- VRAM: 32
- vDisk: 3, first 100 GB, second 250 GB, third 250 GB
What is the easiest way to test VM performance, while minimizing downtime?
Options
- AIncrease the number of vCPUs.
- BMigrate the VM to the AHV cluster.
- CEnable vDisk sharding at AOS level.
- DCollapse the second and the third disk into a single one
How the community answered
(42 responses)- A12% (5)
- B81% (34)
- C2% (1)
- D5% (2)
Explanation
Migrating the VM to the AHV cluster is the easiest way to test whether the Nutanix native storage stack (ADSF/AHV) offers better storage performance than the ESXi-based configuration. Nutanix Move can perform cross-hypervisor migrations with minimal downtime. If performance improves on AHV, it confirms the ESXi storage configuration was the bottleneck. This approach requires no architectural changes to the VM itself. Increasing vCPUs (A) does not address a storage-level issue. Enabling vDisk sharding (C) is a more complex AOS-level operation that requires understanding the workload pattern and is not as straightforward. Collapsing the two 250 GB disks into one (D) requires downtime for data migration and is a structural change that may not improve performance.
Topics
Community Discussion
No community discussion yet for this question.