Submit a ticketCall us

Get a crash course on Network Monitoring delivered right to your inbox
This free 7-day email course provides a primer to the philosophy, theory, and fundamental concepts involved in IT monitoring. Lessons will explain not only how to perform various monitoring tasks, but why and when you should use them. Sign up now.

Home > Success Center > Virtualization Manager (VMAN) > Virtualization Manager (VMAN) Getting Started Guide > Optimize Resources > Evaluate and optimize your VMs

Evaluate and optimize your VMs

Table of contents
Created by Lori Krell_ret, last modified by Melanie Boyd on Nov 22, 2016

Views: 23 Votes: 0 Revisions: 2
VMAN_GS_Home.png

Your virtual machine environment can quickly grow in size and number of resources, becoming unmanageable or difficult to monitor over time. To help manage your environment, Virtualization Manager provides tools to easily evaluate current usage, locate possible allocation issues, and manage resources.

Recommendations help review and manage VMs with predictive actions for pinpointed trends and metrics for VMs that could cause or encounter issues.

VMAN provides additional historical data and metric reporting through dashboards to locate and manage stale, zombie, and rogue VMs. With this continuously polled and collected data, you can better understand and troubleshoot:

  • Active usage of your VMs
  • Allocated resources to over and under utilized systems, requiring balancing
  • Stale, zombie, and rogue VMs costing your company in unused or overused resource

Sprawl dashboard

Virtualization Manager provides a consolidated view through the Sprawl dashboard to help monitor and manage VMs through resources using specific queries and events. The Sprawl dashboard tracks the vital issues that occur in a sprawling environment: resource allocations, snapshots, alerts, and VM resource status. Using the dashboard, you can reclaim wasted VM resources including idle and stale VMs, optimize performance by right-sizing under- and over- allocated VMs, remove orphaned VMDK files, and manage snapshots.

As you review these options, you can click on linked VMs to review additional information. To directly access the VM, use the Virtualization Manager Tools in the node view.

Depending on your configured recommendations, some of these resources may not display data when first installed. You can add additional resources by editing the page.

The Sprawl dashboard includes the following options for reviewing and managing resources:

Resource Description
Top 10 VMs by Underallocated vCPUs

Displays powered on VMs with an average CPU load over 70% in the last 7 days.

These systems may need additional resources, constantly under heavy load and usage.

Top 10 VMs by Overallocated vCPUs

Displays powered on VMs with two or more CPUs with a peak sum of CPU load has not exceeded 30% in the last 7 days.

These systems have resources better balanced and used by under-allocated VMs.

Top 10 VMs by Underallocated Memory

Displays powered on VMs with an average memory consumption over 80% in the last 7 days, ballooning in usage.

These systems may need additional resources, constantly under heavy load and usage.

Top 10 VMs by Overallocated Memory

Displays powered on VMs with 2 or more MBs of memory and the average memory usage has not exceeded 30% in the last 7 days.

These systems have resources better balanced and used by under-allocated VMs.

Top 10 VMs by Snapshot Disk Usage

Displays the latest snapshots taken of the VMs with details on location and size.

You can review these snapshots and delete as needed to reclaim space and clear out older snapshots.

VMs Idle for the Last Week

Displays the VMs idle over the past 7 days with an option to power off the VM. The data includes the average CPU, IOPS, and Net throughput, helping identify zombied VMs.

Idle VMs are not actively in use, best used by under-allocated VMs needing more CPU load, memory, and space.

VMs Powered Off for More Than 30 Days

Displays VMs not powered up or in use for over 30 days. These are stale VMs.

These VMs are lost space, best removed or reallocated to under-allocated VMs needing more CPU load, memory, and space.

VMs that might benefit from decreasing vCPUs

Displays VMs over-allocated with CPUs compared to the tracked load and usage.

Reducing the CPUs can help in costs without degrading performance.

Do not reduce the amount of CPUs below the recommended amounts for the OS.

Orphaned VMDKs

Displays orphaned VMDK files: VMDK files with an associated datastore and without a connected VM.

You can delete the datastore through the resource to directly manage the VMDK files.

If you monitor only a part of your virtual environment in the Orion Web Console, the datastore files that are displayed in this resource might be in use by other, unmonitored hosts. Deleting datastore files that are still in use will irreversibly damage the virtual machine the datastore file belongs to, and you will lose data.

AppStack

You have access to a powerful troubleshooting tool called AppStack. The AppStack displays a visual representation of the entire environment to quickly scan for warning and critical status issues. As you hover over each item, you can review additional status data for at-a-glance insight into issues. Click a monitored node to drill down into additional details pages to further investigate and resolve problems.

Some issues deeper in the environment can bubble up through higher layers. With the visual map of the environment with tooltip status and details, you can clearly pinpoint the originating issue and VM.

Learn more...

Learn more about evaluating and getting the best usage out of your virtual environment:

 
Last modified
18:30, 21 Nov 2016

Tags

Classifications

Public