Zones, Regions and Disaster Recovery

How far apart should copies be?

Cloud availability zones are separate datacentres within a region with independent power and networking; spreading across zones protects against a datacentre-level failure with low latency between copies, and is the standard baseline for production. Multi-region deployments survive the loss of a whole region but are much harder: cross-region latency, data replication, consistency and cost. Two numbers drive the design. RTO (Recovery Time Objective): how long the service may be down. RPO (Recovery Point Objective): how much recent data may be lost. Common disaster recovery strategies, from cheapest to most expensive: backup and restore (hours of RTO, RPO equal to backup interval), pilot light (core data replicated, minimal infrastructure ready to scale up), warm standby (a scaled-down full copy running), and multi-site active-active (near-zero RTO and RPO, highest cost and complexity).

Matching DR strategy to targets

Pick the cheapest option that meets the business's RTO and RPO.

strategy               typical RTO      typical RPO        relative cost
---------------------  ---------------  -----------------  -------------
backup and restore     hours to a day   hours (last backup)   $
pilot light            tens of minutes  minutes              $$
warm standby           minutes          seconds to minutes   $$$
active-active          near zero        near zero            $$$$

ask the business: "If the region vanished now, how long can we be down,
                  and how much data can we lose?"  -> that sets RTO / RPO

Spare keys

A spare key in your drawer (zone redundancy) helps when you lose one key at home. A spare key with a relative in another city (multi-region) helps when the whole house is inaccessible, but it takes longer to fetch and you must remember to update it when you change the lock.

त्वरित जाँच: The business can tolerate 4 hours of downtime and losing up to 24 hours of data for an internal tool. Which DR strategy is most appropriate?

• Multi-site active-active
• Warm standby in three regions
• No backups at all
• Backup and restore