3.4 KiB
Executable File
High Availability, Disaster Recovery and Business Continuity
HA, DR and BC Competency Log
Understanding High Availability (HA), Disaster Recovery (DR), and Business Continuity Planning (BCP) is important when designing reliable systems and analysing service incidents. These concepts help Microlise (and organisations in general) to minimise downtime, recover from failures, and maintain service availability.
High Availability (HA)
High Availability focuses on preventing service disruption by using redundancy. Systems are designed with backup components so that if one fails, another can take over automatically.
Examples include:
- Redundant servers or network paths
- Load balancing across multiple systems
- Automatic failover modes
This ensures services continue operating even when individual components fail. We have this at Microlise where if there is a failure on one of the data centres, we can failover to the other one.
An analogy from the article compares this to a bicycle with two brakes. The bike can operate with only one brake, but having two provides redundancy in case one fails.
Disaster Recovery (DR)
Disaster Recovery focuses on restoring systems after a major failure. This includes recovering infrastructure, applications, and data so services can resume operation.
Some of the common DR tools and methods include:
- Offsite backups
- Replicated environments (e.g., multiple data centres with sync)
- Recovery procedures and restoration tools
Unlike High Availability, which tries to prevent downtime, DR assumes a failure has already occurred. In the bicycle analogy, after the crash the rider takes a bus home and then drives to work to complete their important task. This represents a recovery process after a major disruption.
Business Continuity Planning (BCP)
BCP ensures the business can continue operating during or after a disruption. This may involve alternative systems, temporary processes, or backup locations to keep services running.
In the analogy, the car at home represents BCP because it allows the rider to continue their journey despite the broken bicycle. A Business Impact Assessment (BIA) helps determine the priority of services and the acceptable level of downtime.
RTO and RPO
Two key recovery metrics are:
- Recovery Time Objective (RTO): The maximum time allowed to restore a service after failure. For example, if a critical service must be restored within 4 hours, the disaster recovery process must ensure systems are operational within that timeframe.
- Recovery Point Objective (RPO): The maximum acceptable amount of data loss.
RPO example:
- An RPO of 0 minutes means no data loss is acceptable.
- An RPO of 1 hour means up to one hour of data could be lost.
These values help define the required level of HA and DR design.
Application in Real Scenarios
These concepts are useful when performing incident debugging, root cause analysis, or explaining service issues to customer support teams. For example, engineers may check whether failover worked correctly, whether recovery met the RTO, or whether backups allowed data to be restored within the RPO.
In summary, the reason why we need to understand HA, DR, and BCP is to maintain service availability and ensure business continuity.