The analytics cluster of the vRealize Operations Manager deployment contains the nodes that analyze and store data from the monitored components. You deploy a configuration of the analytics cluster that satisfies the requirements for monitoring the number of virtual machines in the design objectives of this design.

Deployment Type

Deploy a three-node vRealize Operations Manager analytics cluster in the cross-region application virtual network. The analytics cluster consists of one primary node, one primary replica node, and one data node to enable scale out and high availability.

In the design, you deploy the vRealize Operations Manager analytics cluster nodes on the first vSphere cluster in the management domain in Region A. You deploy the vRealize Operations Manager remote collector nodes on the first vSphere cluster in the management domain in the region. With this configuration, you can centrally manage monitoring across the entire SDDC.

You place the vRealize Operations Manager instance in Region A, on a specific virtual network segment. This provides a consistent deployment model for management applications, and supports growth to a dual-region design.

vRealize Operations Manager is distributed as a virtual appliance in OVA format.

To accomplish this design objective, you deploy or reuse the following components to deploy this operations management solution for the SDDC:

  • Cross-region vRealize Suite Lifecycle Manager

  • NSX-T Load Balancer for vRealize Operations Manager

  • Cross-region Workspace ONE Access cluster

  • Supporting infrastructure services, such as Active Directory, DNS, and NTP.

Table 1. Design Decisions on Deployment of vRealize Operations Manager

Decision ID

Design Decision

Design Justification

Design Implication

SDDC-COM-CO-MON-001

Deploy vRealize Operations Manager as a cluster of three nodes - one primary, one primary replica, and one data node, on the first cluster in the management domain in Region A.

  • Provides the scale capacity required for monitoring of up to 10,000 virtual machines.

  • Supports scale-up with additional data nodes.

You must identically size all nodes which increases the resource requirements in the SDDC.

SDDC-COM-CO-MON-002

Deploy two remote collector nodes on the first cluster in the management domain in the region.

Removes the load from the analytics cluster from collecting metrics from region-specific applications.

You must assign a collector group when configuring the monitoring of a solution.

SDDC-COM-CO-MON-003

Use vRealize Suite Lifecycle Manager to deploy vRealize Operations Manager.

Allows vRealize Suite Lifecycle Manager the ability to provide life cycle management of vRealize Operations Manager.

You must deploy vRealize Suite Lifecycle Manager.

SDDC-COM-CO-MON-004

Protect all vRealize Operations Manager nodes by using vSphere High Availability.

Supports the availability objectives for vRealize Operations Manager without requiring manual intervention during a failure event

None

SDDC-COM-CO-MON-005

Apply vSphere Distributed Resource Scheduler (DRS) anti-affinity rules to the vRealize Operations Manager analytics cluster.

Using vSphere DRS prevents the vRealize Operations Manager analytics cluster nodes from running on the same ESXi host and risking the high availability of the cluster.

  • You must perform additional configuration to set up an anti- affinity rule.

  • If additional data nodes are added, you must update the anti-affinity rule.

  • You can put in maintenance mode only a single ESXi host at a time in a management cluster of four ESXi hosts.

SDDC-COM-CO-MON-006

Apply vSphere Distributed Resource Scheduler (DRS) anti-affinity rules to the vRealize Operations Manager remote collector group.

Using vSphere DRS prevents the vRealize Operations Manager remote collector nodes from running on the same ESXi host and risking the high availability of the cluster.

You must perform additional configuration to set up an anti- affinity rule.

SDDC-COM-CO-MON-007

When using two availability zones in Region A, add the vRealize Operations Manager nodes to the primary availability zone VM group, sfo-m01-cl01_primary-az-vmgroup.

Ensures that, by default, the vRealize Operations Manager virtual appliance is powered on within the primary availability zone hosts group.

If vRealize Operations Manager is deployed after the creation of the stretched cluster for management domain availability zones, the VM group for the primary availability zone virtual machines must be updated to include the vRealize Operations Manager nodes.

SDDC-COM-CO-MON-008

Place the cross-region vRealize Operations Manager nodes in a dedicated virtual machine folder in Region A, xreg-m01-fd-vrops.

Provides organization of the vRealize Operations Manager nodes in the management domain inventory

You must create the VM folder.

SDDC-COM-CO-MON-009

Place the region-specific vRealize Operations Manager Remote Collector nodes in a dedicated virtual machine folder in Region A, sfo-m01-fd-vropsrc.

Provides organization of the vRealize Operations Manager Remote Collector nodes in the management domain inventory.

None.

Sizing Compute and Storage Resources

You size resources for vRealize Operations Manager to provide enough resources to accommodate the analytics operations for monitoring the SDDC and the expected number of virtual machines in the SDDC.

Deploying three medium-size nodes satisfies the requirement for retention and for monitoring the expected number of objects and metrics based on the following assumptions.

Table 2. Design Objectives

Number of Virtual Machines

Expected Number of Data Center Objects

10,000

12,500

You allocate storage capacity for analytics data collected from the management products and from the number of tenant virtual machines that are defined in the objectives of this SDDC design.

This design uses medium-size nodes for the analytics cluster and standard-size nodes for the remote collector group. To collect the required number of metrics, you must add a virtual disk with the size of 1 TB to each analytics cluster node.

Table 3. vRealize Operations Manager Analytics Cluster CPU, Memory, and Storage Resources

Attribute

Per Appliance

Cluster Deployment

Appliance size

Medium

-

vCPUs

8

24

Memory

32 GB

96 GB

Initial Storage

274 GB

822 GB

Additional Storage

1 TB

3 TB

Unlike the analytics cluster nodes, remote collector nodes have only the collector role. Deploying two remote collector nodes in the region does not increase the capacity for monitored objects.
Table 4. vRealize Operations Manager Remote Collector CPU, Memory, and Storage Resources

Attribute

Appliance

Appliance size

Standard

CPU

2 vCPUs

Memory

4 GB

Additional Storage

N/A

Table 5. Design Decisions on Sizing of vRealize Operations Manager

Decision ID

Design Decision

Design Justification

Design Implication

SDDC-COM-CO-MON-010

Deploy each node in the analytics cluster as a medium-size appliance.

  • If you use fewer large-size vRealize Operations Manager nodes, you must increase the minimum host memory size to handle the increased performance that is the result from stretching NUMA node boundaries.

  • Provides enough capacity for the metrics and objects generated by up to 12,500 objects while having high availability in the analytics cluster enabled. Metrics are collected from the following components:

    • vCenter Server instances

    • ESXi hosts

    • NSX-T Data Center components

    • vRealize Automation

    • vRealize Log Insight

    • Storage Array and data center infrastructure

ESXi hosts in the management cluster must have physical CPUs with a minimum of 8 cores per socket. In total, vRealize Operations Manager uses 24 vCPUs and 96 GB of memory in the management cluster.

When you exceed 12,500 objects, you must deploy additional nodes.

SDDC-COM-CO-MON-011

If the number of SDDC objects exceeds 12,500, add more medium-size nodes to the analytics cluster.

Ensures that the analytics cluster has enough capacity to meet the SDDC object and metric growth.

  • The capacity of the physical ESXi hosts must be enough to accommodate virtual machines that require 32 GB RAM without bridging NUMA node boundaries.

  • The management cluster must have enough ESXi hosts so that vRealize Operations Manager can run according to vSphere DRS anti-affinity rules.

  • The number of nodes must not exceed number_of_ESXi_hosts in_the_management_cluster - 1.

    For example, if the management cluster contains six ESXi hosts, you can deploy up to five vRealize Operations Manager nodes in the analytics cluster.

SDDC-COM-CO-MON-012

Increase the initial storage of each vRealize Operations Manager analytics cluster node by 1 TB.

Supports the storage requirements for monitoring up to 12,500 objects.

None

SDDC-COM-CO-MON-013

Deploy each remote collector node as a standard-size appliance.

  • Enables metric collection for the expected number of objects in the SDDC when at full capacity.

  • Remote collectors do not perform analytics operations or store data on disk, therefore no additional storage is required.

You must provide 4 vCPUs and 8 GB of memory in the management cluster in Region A.