Search <book_title>...

Veritas Data Insight Administrator's Guide

Last Published: 2020-07-14

Product(s): Data Insight (6.1.5)

Understanding Data Insight best practices

The following topics describe the common practices and sizing guidelines:

Common practices
General sizing guidelines

Common practices

Use 64-bit server architecture for better Data Insight performance.
Use high performance disks for indexers, SAN disk for ease of expansion and backup
- Disk concatenation or RAID 4 may generate hotspots in the array subsystem. Ensure proper distribution to avoid hotspots.
Use separate disks for Install and Data directories.
Assign faster and bigger Temp directory for indexers.
- Temp directory (for example, in Windows, the directories that are indicated by TMP and TEMP system environment variables) must be on disk other than system disk (operating system disk). This best practice improves index vacuum (like defrag for segments) performance, and overall performance.
Assign indexer memory of 1 GB per thread.
Assign indexer memory sizing of 1 GB per 10 million audit events.
Assign low latency high bandwidth (or dedicated) network links for Collector-Filer communication.
Create exclude rules to ensure that the information that is reported is relevant and useful.
- If a third-party application that generates a lot of events, resides on a volume, exclude that volume from auditing (like NetApp FPolicy) to restrict events.
Schedule scans at off peak hours to minimize effect on users accessing the shares.
Disable scans from running during peak usage.
Manage job scheduling as per load.
Assign appropriate roles and access for Data Insight users.
Deploy DFS mappings to ensure that all names are recognizable and meaningful.
Create containers to logically group related objects together for administration and reporting purposes.
Set up event notifications to ensure that errors and warnings are reported.
Define retention policies to ensure the database and log files are maintained over time.
Define report retention in when configuring reports ("Reports to preserve" setting).
Maintain adequate proximity - Management Server close to indexers, Collector close to filers.
Use Chrome for faster UI.
Refer to the Veritas Data Insight Installation Guide for the latest platform and version support information.

General sizing guidelines

The following sizing guidelines are typically recommended for Data Insight setups. The recommendations may change based on environment.

Management Server
- Up to 16 indexers
Indexers
- Up to 20,000 shares
- Allocate 100 MB of disk space per million files
- Allocate 20 MB of disk space per million events
- Typical storage size range is 40 GB to 400 GB depending upon environment. Large deployments may even need more space.
Collectors
- Up to 10,000 shares or ten filers, whichever comes first
- Up to 150 Windows NAS agents
- Up to 20 SharePoint web front-end servers
- Typical storage size range is 60-80 GB per collector. Additional space may also help in case of network outages where data is staged before it is transferred to Indexer.
Indexer storage
- 100 MB/1 million files
- 20 MB/1 million events
Scan times
- 200 files per second per thread

Risk Dossier sizing guidelines

If you enable the computation of the Risk Dossier, there will be an overall increase in the space required for the index folder.

For the Management Server, the dossier database size can be calculated as:

Number of users * 2 MB (to store dossier date for 6 months)

Thus, for 50000 users, provision up to 100 GB space to store dossier data of 6 months.