Chat now with support
Chat with Support

Enterprise Reporter 3.2.2 - Installation and Deployment Guide

Product Overview Installation Considerations for Enterprise Reporter Installing and Configuring Enterprise Reporter Managing Your Enterprise Reporter Deployment Troubleshooting Issues with Enterprise Reporter Appendix: Database Content Wizard Appendix: Encryption Key Manager Appendix: Log Viewer

Cluster Deployment Considerations

The following sections can help you get the most out of Enterprise Reporter:

Consider the Data to Collect Before Deploying Nodes

There is a relationship between the design of your discoveries and the physical deployment of your clusters and nodes. By understanding what you want to collect, and what network and hardware resources you have available, you can deploy Enterprise Reporter in a way that makes sense for your environment. When you create a discovery, you assign it to a cluster. The actual work of the collection can be done by any node in the cluster, load balanced by the Enterprise Reporter server. This means that each node should:

You may not have this insight before you start using Enterprise Reporter, however you can modify your deployment at any time. If you are creating a discovery with targets that are not near the nodes of any of your clusters, consider creating a new cluster to support this.

Topics:  

Fine Tune Each Cluster and Node

A discovery is resolved to a number of tasks. Each task is then automatically assigned to a node in the cluster by the server, depending on the node’s availability. If you are collecting from many targets at the same time, you can increase performance by adding more nodes to the cluster and by ensuring that each node is configured to allow the node to optimize how many concurrent tasks it can process (by setting the maximum number of concurrent tasks to a value of zero).

You can see the amount of time it took to run each instance of the discovery in the history view. If you drill down, you can see how the time was distributed across the targets, and what node did the processing. This information can aid in your decisions about how to scale up your deployment.

Adding nodes and optimizing concurrent tasks can only speed things up if there are multiple tasks to assign to the node. When a large amount of data is collected from an individual target, only one task will be created. In that case, you can improve performance by increasing the CPU, the available disk space, and the memory of the node host computer, or by looking at other factors, such as network latency.

The discovery type determines what the targets are and helps you decide on your options for speeding up collections. Clusters that handle large targets will benefit from increasing the CPU, available disk space and memory of the node host computers, or being dedicated to a smaller number of discoveries. The following table outlines how each discovery type is broken down into tasks:

Table 17. Discovery Types

Active Directory®

Each domain, or per object type by domain

Azure Active Directory

Each tenant

Azure Resource

Each tenant

Computer

Each computer

Exchange®

The Exchange Organization, or per object type, or per object type per server

Exchange Online™

The Office 365 tenant

File Server Analysis

Each computer

Microsoft SQL

Each computer

Microsoft Teams

Each tenant

NTFS

Each computer or per share by computer

OneDrive

Each tenant

Registry

Each computer

Topics:  

Optimize Node Setup

Two of the most common questions are, “What nodes do I deploy?’ and, “What computer specifications do I need?” To determine the answers to these questions, we typically look at the following criteria.

The following sections outline considerations when optimizing node setup for each type of discovery.

Active Directory collections are sequential and create heavy network traffic as they query the domain controller so the network connection to the domain controller is the primary concern. Locating the node close to the domain controller is recommended. Choose a domain controller close to your node when you configure the discovery. Technical Documentation.

For more information, see Choosing your Active Directory Scopes.

The CPU benchmarks of the node computer affect the threading capability so it is a secondary concern.

Typically, an organization with one domain only needs one node. Additional nodes usually only help optimize concurrent collection when there are multiple domains.

 

Small

< 100K
1 domain

Primary
Concern

No
Concern

No
Concern

Secondary
Concern

1

 

Medium

100K - 500K
1 domain

Primary
Concern

No
Concern

No
Concern

Secondary
Concern

 

1

 

Large

500K - 1M
1 domain

Primary
Concern

No
Concern

No
Concern

Secondary
Concern

 

1

•break single discoveries by object type

•use one discovery per object type combined with schedules

Azure Active Directory collections are sequential and create heavy network traffic as they query Azure. Ensuring the node machine has optimal network bandwidth is the primary concern. The CPU benchmarks of the node computer affect the threading capability, so it is a secondary concern. Typically, an organization with one tenant only needs one node.

 

Small

< 100K
1 tenant

Primary
Concern

No
Concern

Tertiary
Concern

Secondary
Concern

1

Medium

100K - 500K
1 tenant

Primary
Concern

No
Concern

Tertiary
Concern

Secondary
Concern

1

Large

500K - 1M
1 tenant

Primary
Concern

No
Concern

Tertiary
Concern

Secondary
Concern

1

Computer collections are sequential and create heavy network traffic as they query each local computer. Ensuring the node machine has optimal network bandwidth is the primary concern.

 

Small

< 5K

Primary
Concern

No
Concern

Secondary
Concern

Tertiary
Concern

1 - 3

Medium

5K - 10K

Primary
Concern

No
Concern

Secondary
Concern

Tertiary
Concern

3 - 8

Large

> 10K

Primary
Concern

No
Concern

Secondary
Concern

Tertiary
Concern

8 - 10

 

Small
1 - 10
Computers

< 2M

Primary
Concern

Secondary
Concern

No
Concern

Secondary
Concern

1 - 3

Medium

1 - 10
Computers

< 20M

Primary
Concern

Secondary
Concern

No
Concern

Secondary
Concern

3 - 9

Large

> 10
Computers

> 20M

Primary
Concern

Secondary
Concern

No
Concern

Secondary
Concern

10

 

 

Small

1 - 3

Primary
Concern

Tertiary
Concern

Secondary
Concern

No
Concern

1

Medium

3 - 5

Secondary
Concern

Primary
Concern

Tertiary
Concern

No
Concern

2

Large

> 5

Secondary
Concern

Primary
Concern

Tertiary
Concern

No
Concern

> 3

 

The most important guideline is to collect only the information required. For example, most files have inherited permissions so, typically, collecting folder permissions is sufficient.

By default, NTFS discoveries, will create multiple tasks (one task per share) to improve performance. If disk speed is slow, network bandwidth is low, or there is only one node, disable this performance option.

 

Small

0 - 5M

Secondary
Concern

Tertiary
Concern

Primary
Concern

No
Concern

1 - 3

 

Medium

5M - 100M

Secondary
Concern

Tertiary
Concern

Primary
Concern

No
Concern

3 - 6

 

Large

100M - 1B
multiple shares

Secondary
Concern

Tertiary
Concern

Primary
Concern

No
Concern

6 - 101

•use multiple tasks option unless slow disk speed, low network bandwidth, or one node


1

These considerations apply to Exchange Online, Microsoft Teams, and OneDrive discoveries.

OneDrive can be divided into multiple discoveries to increase collection speed. If Microsoft throttling is often an issue, the use of multiple credentials can help minimize throttling.

 

Small

 

Primary
Concern

No
Concern

Tertiary
Concern

Secondary
Concern

1

Medium

 

Primary
Concern

No
Concern

Tertiary
Concern

Secondary
Concern

1

Large

 

Primary
Concern

No
Concern

Tertiary
Concern

Secondary
Concern

1

 

Related Documents

The document was helpful.

Select Rating

I easily found the information I needed.

Select Rating