Disaster Recovery Flashcards
1
Q
RTO
A
Recovery Time Objective
- Period of time system can remain offline after disaster
2
Q
WRT
A
Work Recovery Time
- Additional work for testing and integration to restore for full business function
3
Q
MTD
A
Maximum Tolerable Downtime
- Max time a system can be down
4
Q
RPO
A
Recovery Point Objective
- The amount of data loss a system can sustain measured in time units
- For instance, if a database was destroyed RPO of 24 means it can be recovered from not more than 24 hours before hand
5
Q
MTBF
A
Mean Time Between Failures
- Life expectancy of product
- Total operational time divided by number of failures
6
Q
MTTF
A
Mean Time to Failure
- Averaged time til a non replaceable component fails
- Used to calculate needed redundancy necessary
7
Q
MTTR
A
Mean Time to Repair
- Measure of time taken to correct a fault
- Could be time taken to replace or restore as well
- total number of hours of unplanned maintenance divided by number of failures
8
Q
PDU
A
Power Distribution Unit
- “Cleans” a power signal, providing protection from spikes and drops.
- Integrates with UPS
9
Q
Multi-pathing
A
- Network node has more than one physical link to another node
- Default for full and partial mesh
10
Q
SAN Multipathing
A
- Each server has at least two SAN controllers each with a dedicated link to the storage network
11
Q
Multiple ISPs
A
- Multiple WAN hookups from different ISPs.
- Must make sure they are using separate infrastructure
- Diverse paths of entry increases redundancy
12
Q
NIC Teaming
A
- Combining two or more cabled connections to make one logical channel
- AKA Link Aggregation
- Also called bonding
- Provides redundancy so long as the function does not demand full speed of bonded link
13
Q
Load Balancer
A
- Distributes server requests over nodes in a farm or pool
- Layer 4 Switch,
- Layer 7 Switch (content switch), makes decisions based on content based data. Requires a mode complex processor.
14
Q
Clustering
A
- Allows multiple processing nodes that share data
- If one stops working, it fails over to others
15
Q
Virtual IP
A
Shared or Floating address
- Public IP is shared between two instances in a cluster
- Active node owns the IP
- Fails over automatically based on heartbeat