Definition
Unplanned downtime is caused by hardware, software or administrator
failure that was not "protected" by the system. These failures cause
an unanticipated denial of service to users and applications.
Customer problem
Today's NonStop e-business environment does not leave room
for unplanned downtime. Systems, applications and data must be available
to users 100% of the time to meet daunting customer service levels.
HP's answer
Tru64 UNIX 's evolutionary reliability and availability architecture
and design are a strong statement to you that hp is committed to solving
your availability and performance needs by providing high availability
leadership solutions. HP's goal is to minimize unplanned downtime so
that system availability stays at the highest possible level. Here's
how:
TruCluster Server: No matter
how reliable your single system is, only a clustered configuration
can keep your business running absolutely uninterrupted. TruCluster
Server software guarantees your service levels with maximum availability
while reducing your management costs. How? By managing all systems
in the cluster as one single system. No other vendor can offer you
this level of availability and ease of management at the same time.
Alternate pathing: Tru64 UNIX
supports alternate pathing with automatic failover (transparent to
applications) for storage, network peripherals, and cluster interconnects.
In a cluster environment, DRD provides failover to storage through
other systems in the cluster; essentially providing alternate pathing
at the system level. The LSM software (Logical Storage Manager) also
provides transparent failover to mirror sets.
Automatically remove failing components:
Tru64 UNIX serviceability features allow the system to automatically
detect failing components such as CPU, memory, or environmental devices,
and shut them down to avoid system crashes. You can therefore experience
maximum system availability until you are ready to service these components
at your most convenient time.
Memory error resiliency:
Tru64 UNIX guards memory with a "just-in time" scrubbing model, where
correctable errors are scrubbed when encountered by the operating system
or application code, ensuring maximum availability. To enhance this
capability, Tru64 UNIX has implemented a trigger mechanism, called
the memory troller, which pro-actively locates and scrubs correctable
memory errors prior to the operating system or application reading
the memory. The troller arranges for the memory page containing the
error to be marked bad and if the page is free (or when it becomes
free) maps the page out, so it will not be reused.
Protect your systems from disaster: Tru64
UNIX solutions protect IT systems from disaster by replicating systems
at multiple sites to ensure continuous operation in the event of a
disaster (ie fire, flood, earthquake, etc.) There are several ways
to achieve degrees of disaster protection, from backups to extended
clusters to more complex configurations. To learn more about how to
enhance high availability and include disaster protection, click
here. This white paper includes product information on Campus-Wide
Disaster Tolerant Clusters and Data
Replication Manager, key components of a robust disaster protected
environment. Additionally, information is included on Enterprise
Volume Manager and Reliable
Transaction Router products that enhance the IT environment that
requires high availability and disaster protection.
|
