Home › Topics › Retention and Granularity: How Long to Keep What

Retention and Granularity: How Long to Keep What

Every retention decision is a negotiation between the person doing forensics, the auditor, and the disk. These guides make the negotiation concrete. They explain what each stakeholder actually needs and the arithmetic that tells you what your data will cost in storage. They cover designing tiers from raw to hourly to daily to gone, and keeping the monitoring database itself healthy. They also cover archiving old data in a way you can prove you can get back.

Guides in this topic

  1. Forensics, Auditors, and the Disk: Three Stakeholders in Every Retention DecisionWhat each one needs, for how long, at what resolution; where they conflict; how to settle it; and a retention matrix template.
  2. Storage Math: What Your Monitoring Data Will Cost in DiskThe formula is sources times metrics times samples per day times bytes times days. It is worked for metrics, logs, and flows, with a script that estimates yours.
  3. Designing Retention Tiers: Raw, Hourly, Daily, and GoneAging schedules, what survives each rollup, why an average hides the spike that mattered, keeping maximums, and a tier table you can adopt.
  4. Monitoring Database Housekeeping: Purging, Compacting, and Not Filling the DiskPurge jobs, reclaiming space, index bloat, free-space alarms on the monitoring server itself, backing up the monitoring database, and checking that housekeeping actually runs.
  5. Archiving Old Monitoring Data (and Proving You Can Get It Back)Export formats, cold storage, what to keep forever, the annual restore test, and the incident snapshots that should never be rolled up.

Put this into practice with SysPeep Monitor

SysPeep is network and infrastructure monitoring software for Windows. It discovers the devices on your network, polls their health and performance over SNMP and WMI, and alerts you the moment something goes wrong — all from one web console.

Download Free TrialLearn About SysPeep