I thought I would share a quick demonstration showing how Nutanix Acropolis Hypervisor (AHV) load balances VMs at power on to ensure optimal performance.
This is just one of the many seemingly simple functions of AHV to ensure optimal performance throughout the cluster. Its pretty simple, when AHV issues a Power On request it queries the Acropolis Master which continually receives compute (CPU/RAM) and Storage (Capacity/Performance) data from the Acropolis Slaves (other nodes) in the cluster.
When the Acropolis Master receives the power on request it simply looks for the node with the lowest utilization and issues the command to power on the VM.
This is a simple, proactive way of minimising the chance of compute contention within a cluster.
This is Part 4 of the Jetstress performance testing on Nutanix Acropolis Hypervisor (AHV) series of videos.
This video shows the following:
Jetstress performance on the forth node in the cluster after the VM has already been migrated in quick succession from its original node too a 2nd and 3rd node both of which have ran a Jetstress test.
Nutanix Controller VM (CVM) CPU utilization
This test is the forth of several test to show the impact to performance by migrating VMs around several nodes in the same cluster which results in not all data being accessible/stored locally.
These post migration tests are designed to show the worst case performance when business critical workloads with large data sets (such as Exchange) move around nodes within a Nutanix AHV cluster. By running Jetstress on multiple nodes in quick succession, the workload on the Nutanix Controller VMs (CVMs) to localize data and maintain a even balance of data across all drives (SSD/HDD) and nodes within the cluster is higher than normal.
While migrating business critical applications around multiple nodes in quick succession (especially while running intensive benchmarks at much higher IOPS that the real world requires) is not something that will occur (unless initiated by an administrator), it demonstrates the performance impact (or lack thereof) of even this unrealistic operation.
Note: As with previous videos, This demonstration is not showing the peak performance which can be achieved by Jetstress on Nutanix.
This is Part 3 of the Jetstress performance testing on Nutanix Acropolis Hypervisor (AHV) series of videos.
This video shows the following:
Jetstress performance after the VM has been migrated in quick succession from its original node to a 2nd node, which runs a Jetstress test and then the performance of the VM on a third node within the same NDSF cluster.
This test is the second of several test to show the impact to performance by migrating VMs around several nodes in the same cluster which results in not all data being accessible/stored locally.
These post migration tests are designed to show the worst case performance when business critical workloads with large data sets (such as Exchange) move around nodes within a Nutanix AHV cluster. By running Jetstress on multiple nodes in quick succession, the workload on the Nutanix Controller VMs (CVMs) to localize data and maintain a even balance of data across all drives (SSD/HDD) and nodes within the cluster is higher than normal.
While migrating business critical applications around multiple nodes in quick succession (especially while running intensive benchmarks at much higher IOPS that the real world requires) is not something that will occur (unless initiated by an administrator), it demonstrates the performance impact (or lack thereof) of even this unrealistic operation.
Note: As with previous videos, This demonstration is not showing the peak performance which can be achieved by Jetstress on Nutanix.
