iwantcoding.com
🔥 Daily 👥 Rooms 🏆 Top Log in Sign up

10.1 DR Planning: RTO & RPO

Disaster Recovery Planning defines how fast you must come back (RTO) and how much data you can afford to lose (RPO). The worked example computes a budget allocation across tiers.

10.1 Disaster Recovery Planning - RTO and RPO

RTO and RPO on a timeline

Last good backup Disaster! Service restored RPO window (data loss) RTO window (downtime) Figure 10.1 - RPO = data lost between backup and disaster; RTO = downtime between disaster and recovery.

Strategy options by RTO target

RTO targetStrategyCost
DaysBackup + restore from tape / objectLow
HoursPilot light (core infra warm)Low-med
MinutesWarm standbyMedium
SecondsActive-active multi-regionHigh

RPO is bounded by replication

MechanismTypical RPO
Periodic backuphourly or daily
Async replicationseconds to minutes
Sync replicationnear zero (latency cost)
Cross-region activenear zero with conflicts

Worked example - bank tiered DR budget

TierSystemsRTORPOStrategyBudget / yr
Tier 1Core banking1 hr15 minCross-region syncPHP 80m
Tier 2Customer portals4 hr1 hrWarm standbyPHP 25m
Tier 3Internal apps1 day1 dayBackup + restorePHP 5m
Total~PHP 110m

DR test cadence

Test typeCadence
Tabletop (discussion)Quarterly
Functional restoreEvery 6 months
Full failoverAnnual

A plan that has never been tested is not a plan; it is a hope.

Philippine context

  • BSP guidance mandates tested DR for BSFIs.
  • Insurance Commission similar for insurers.
  • Typhoon-related regional outages common; multi-region matters.
Mentor’s tip: RTO from business; RPO from data. Tiered DR budgets aligned to tiers of system criticality. Test or it is hope, not a plan - tabletop quarterly, functional every 6 months, full failover annually.

Discussion

Loading…