3.4 Resilience and recovery

High Availability

Ensures services remain accessible during disruptions

Example: Using multiple web servers with a load balancer

Load Balancing (HA)

Distributes traffic across multiple systems

Example: Round robin load balancer for web traffic

Clustering (HA)

Multiple servers work as one to ensure failover

Example: Database cluster with a automatic role reassignment, adjusting and taking over if one fails

Hot site (site consideration)

Fully operational backup site, ready instantly

Example: duplicate data center available 24/7

Warm Site (site consideration)

Partially equipped site, requires some setup

Example: Servers and power exists but no live data

Cold Site (site consideration)

Empty site with infrastructure only

Example: Office space with power but non servers

Geographic Dispersion (site consideration)

Placing systems in different regions

Example: Datas in US an EU to prevent regional outage impact

Platform Diversity

Using varied operating systems or hardware to reduce common vulnerabilities

Example: Mixing Windows, Linux and Mac endpoints

Multi Cloud Systems

Using more than one cloud provider for redundancy

Example: Hosting critical services on both AWS and Azure

Continuity of Operations

Ensures business processes continue during incidents

Example: Activating remote work plan after office flood

People

(Capacity Planning)

Ensures staffing levels can support business needs

Example: On-call IT rotation during holidays

Technology

(Capacity Planning)

Ensures systems can meet capacity demand
Example: Provisioning extra servers ahead of product launch

Infrastructure

(Capacity Planning)

Ensures sufficient power, cooling, etc

Example: Adding HVAC units as server racks increases

Tabletop Exercises (Testing)

Discussion based test of incident response plans

Example: Walkthrough of ransomware response scenario

Failover (Testing)

Switching to standby system when primary fails


Example: Automatic switch to backup ISP during outage

Simulation (Testing)

Real time emulation of failures

Example: Simulating power outage for backup system test

Parallel Processing

(Testing)

Running systems in sync to take over instantly

Example: Two mirrored DBs processing same transactions

Onsite/Offsite (Backup)

Copies of data stored locally and externally

Example: Backups stored on local NAS, network attached storage, and cloud storage

Frequency (Backup)

How often backups are taken

Example: Hourly,daily or weekly backups

Encryption (Backup)

Protecting backups from unauthorized access

Example: Encrypting tape backups before storage

Snapshots (Backup)

Point in time data copies

Example: VM snapshot before software upgrade

Recovery (Backup)

Restoring systems and data to working state

Example: Recovering a database after corruption

Replication (Backup)

Real time data duplication to another location

Example: Synchronous database replication to secondary site

Journaling (Backup)

Log changes to track and recover data

Example: Synchronous database replication to secondary site

Generators (power)

Backup power for long outages

Example: Diesel generator starts during blackout

UPS (Uninterruptible power supply) (Backup)

Short term power to allow graceful shutdown

Example: UPS keeps firewall online during powercut