CompTIA Security+ Preparation · lesson 26 of 45 · 25 min
High Availability and Redundancy
High availability (HA) — designing systems to minimize downtime. Achieved through redundancy, failover, and load balancing. Measured as uptime percentage: 99.9% = 8.7 hours downtime/year; 99.99% = 52 minutes/year; 99.999% = 5 minutes/year.
Clustering — multiple servers work together as one. Active-active: all nodes handle traffic. Active-passive: standby node takes over on failure.
RAID (Redundant Array of Independent Disks) — combines multiple disks for performance and/or redundancy.
RAID 0: striping — performance, no redundancy. RAID 1: mirroring — full redundancy, 50% capacity. RAID 5: striping with parity — can survive one disk failure. RAID 6: two parity disks — survives two failures. RAID 10: mirror + stripe — best performance + redundancy.
Geographic redundancy — data centers in multiple locations. Protects against regional disasters.
Backup Strategies and Recovery Planning
Backup types: Full (all data, slowest to back up, fastest to restore), Incremental (only changes since last backup — fastest to back up, slowest to restore), Differential (changes since last full — medium speed both ways).
3-2-1 rule — 3 copies of data, on 2 different media types, with 1 copy offsite.
RTO (Recovery Time Objective) — maximum acceptable downtime. How quickly must systems be restored?
RPO (Recovery Point Objective) — maximum acceptable data loss. How much data can you afford to lose?
BCP (Business Continuity Plan) — how the organization continues operating during a disruption. Broader than IT — covers people, processes, facilities.
DRP (Disaster Recovery Plan) — specific IT procedures for recovering systems after a disaster. Subset of BCP.
Site types: Hot site (fully operational, immediate failover, most expensive), Warm site (partially equipped, hours to days to activate), Cold site (empty facility, weeks to activate, cheapest).
Key Terms
Key Terms
Knowledge Check — 10 Questions
Security+ — Resilience & Recovery
RAID 1 provides fault tolerance through:
- Striping
- Parity
- Mirroring — identical data on two disks
- Compression
- Answer: Mirroring — identical data on two disks
RAID 1 writes identical data to two disks — if one fails, the other has a complete copy. 50% storage efficiency. RPO defines:
- How quickly systems must be restored
- How much data loss is acceptable
- The cost of recovery
- The number of backup copies
- Answer: How much data loss is acceptable
RPO = Recovery Point Objective — the maximum age of data that can be lost. If RPO is 4 hours, backups must run at least every 4 hours. The 3-2-1 backup rule requires:
- 3 full backups weekly
- 3 copies, 2 media types, 1 offsite
- 3 sites, 2 servers, 1 cloud
- 3 daily, 2 weekly, 1 monthly
- Answer: 3 copies, 2 media types, 1 offsite
3-2-1: 3 copies (original + 2 backups), on 2 different media (disk + tape, or disk + cloud), with 1 copy offsite for disaster recovery. A hot site differs from a cold site because:
- Hot sites are cheaper
- Hot sites are fully operational and provide immediate failover
- Cold sites are faster to activate
- They are identical
- Answer: Hot sites are fully operational and provide immediate failover
Hot site: fully equipped and running — failover in minutes. Cold site: empty building — weeks to activate. Hot sites cost much more.
Which backup type is fastest to back up but slowest to restore?
- Full
- Differential
- Incremental
- Mirror
Answer: Incremental
Incremental backs up only changes since the last backup — fast to back up, but restoration requires the last full + all incrementals.
RAID 0 provides:
- Fault tolerance
- Performance through striping, but no redundancy
- Both performance and redundancy
- Parity-based recovery
- Answer: Performance through striping, but no redundancy
RAID 0 stripes data across disks for speed — if any disk fails, ALL data is lost. No fault tolerance. BCP is broader than DRP because:
- BCP only covers IT systems
- BCP covers people, processes, and facilities — not just IT recovery
- DRP is more comprehensive
- They are the same
- Answer: BCP covers people, processes, and facilities — not just IT recovery
BCP = how the whole business continues operating. DRP = how IT systems are recovered. DRP is a subset of BCP. Active-active clustering means:
- One node is always on standby
- All nodes handle traffic simultaneously
- Nodes alternate weekly
- Only one node is active at a time
- Answer: All nodes handle traffic simultaneously
Active-active: all cluster nodes handle production traffic — provides both load balancing and failover.
RAID 5 requires a minimum of how many disks?
- 2
- 3
- 4
- 5
Answer: 3
RAID 5 requires at least 3 disks — data and parity are striped across all disks, allowing recovery from one disk failure.
RTO stands for Recovery _____ Objective.
Answer: Time
RTO = Recovery Time Objective — the maximum acceptable time to restore systems after a failure.
Previous lesson: Zero trust architecture: identity, microsegmentation, least privilege
Back to CompTIA Security+ Preparation
Next lesson: Infrastructure as code (IaC) and secure DevOps: CI/CD pipeline security, SAST, DAST