A-27
Theory of Controller Operation
■
allows for the nodes to be geographically located?
How Available are Clusters?
outlines the maximum availability per downtime in the
different architectures:
Application of Availability
The DuraStor storage subsystem implementation of availability
within its supported topologies are shown in
.
Table A-7 Maximum Availability Per Downtime
Architecture
Maximum
Availability
Downtime per
Failure
Downtime per Year
(in minutes)
Continuous
Processing
100.00%
None
0
Fault-Tolerant
99.9999%
Cycles
0.5 – 5
Clusters
99.9 – 99.999%
Seconds to
minutes
5 – 500
High
Availability
99.9%
Minutes
500 – 10,000
(disk mirroring)
Stand Alone
System
99.5%
Hours
2,600 – 10, 000
(without disk
mirroring)
Table A-8 Implementation in Topologies
Architecture
Corresponding Topology
Continuous Processing
Not Available
Fault-Tolerant
Active-Passive Dual-Port
Cluster
Active-Active Dual-Port
High Availability
Active-Active Single-Port
Stand Alone System
Stand-Alone Dual-Port
DuraStor IUG - Online.book Page 27 Friday, July 19, 2002 3:06 PM