Disaster recovery map of a regional airline

Views in this map

1. 1
   Normal runningEast US 2 runs every system, Central US a warm copy. Tier 0 must be back in 15 min.
2. 2
   What replicatesHow each copy gets its data: tier 0 risks 5 minutes. Reporting and BI have no copy.
3. 3
   Failover rehearsal13 September, 06:00: East US 2 is lost. The first release comes at 06:22, 7 min late.
4. 4
   What it foundFive gaps the rehearsal found, each pinned beside the copy it stopped.
5. 5
   After the fixesEvery gap closed. On 18 October the first flight is released in 11 minutes.

Normal running ‹ › What replicates ‹ › Failover rehearsal ‹ › What it found ‹ › After the fixes

Flows to play

• ▷
  Releasing a flightFlight 1102, Raleigh–Durham to Boston at 06:50: every tier 0 system has a part in its release. In the Normal running view.
• ▷
  Losing the primary regionThe rehearsal of 13 September, from declaring East US 2 lost to the first flight released from Central US. In the Failover rehearsal view.
• ▷
  The second rehearsalThe same rehearsal on 18 October, with the five fixes in place. In the After the fixes view.

The model behind it

63 records of 8 types. Every box on the map is one of them, so the filters and the record panel read the same data.

• Systems 26
• Replications 12
• Runbook steps 6
• Gaps 5
• Fixes 5
• Regions and sites 4
• Vendor services 3
• Rehearsals 2

Filters

Tier and Region or site and Owner: pick a value and everything else fades.