When focused on anticipating, troubleshooting, and preventing poor user experience, only the most current data is relevant. What happened last month or even a year ago has no bearing on what Goliath Performance Monitor (GPM) was built for. To ensure you have the data you need to troubleshoot user experience issues, Goliath recommends you keep at least 30 days of data.
How GPM Manages Data Retention
GPM manages data retention through a database Auto Purge feature, which can be set to run at a particular time each day. Depending on the metric or object, you can retain data up to a specific number of days as detailed below.
The following metrics can be stored up to 180 days:
- XA/XD/Horizon (Session & Performance Metrics)
- Server Metrics (Hypervisor Host/VM/Storage Metrics)
- Counter Watch (Windows Perfmon Metrics)
- Server Watch (PING, AGENT, Hypervisor Availability Metrics)
- Event Log Watch (Windows Event Log Data)
- Syslog Watch (Syslog Message Data)
- All Other monitoring data
Non-metric data retention:
- Report Files (up to 550 days)
- Sever Log files (up to 60 days)
- VMware Hypervisor Obsolete Inventory (up to 550 days)
- XenServer Hypervisor Obsolete Inventory (up to 550 days)
WARNING: Keeping large amount of data will degrade overall product performance (e.g. storing data, reporting, screens)
To access the Auto Purge feature
The Auto Purge feature is accessed by going to Settings > Database Maintenance. The Database Maintenance dialog is displayed below.