White paper (Introduction)
Management with HACMP/XD and PPRC
The success of your business depends on the availability of your applications,
networks, and data. To thrive in today's competitive market, information
must flow freely to wherever it is needed in the company, whenever it
is needed. If your profit depends on real-time online processing systems,
then you are vulnerable to a profound negative impact should a catastrophe
cripple your data center. A catastrophe could be as dramatic as an earthquake
or a hurricane, or as mundane as a power failure, water main break, or
loss of a network or storage adapter. Whatever the cause, every minute
of downtime could cost you lost revenue, lost opportunity, and even customer
must be able to respond quickly to restore essential services when a disaster
strikes. Maintaining backup tapes at another site is not enough. You face
an unknown amount of downtime before you can reestablish service, and
chances are that the data you restore is already out of date. Meanwhile,
your employees are idle, your customers worry, and you lose business.
What you need is a solution that ensures that critical data and applications
remain available, no matter what.
Your Business from The Cost of Downtime
IBM addresses your need for continuous access to your system with a suite
of hardware and software products designed to maintain high availability
of your applications, data, servers, and storage. A disaster recovery
plan that includes geographically dispersed Enterprise Storage Servers
(ESS), Peer to Peer Remote Copy (PPRC) and High Availability Cluster Multi-Processing
eXtended Distance for AIX (HACMP/XD) will automate the real-time replication
of your applications and data while monitoring and maintaining the availability
of your system. These three products work together as an integrated system
- Automatic backup
and recovery after failures - The recovery of your business-critical
applications and data after a wide range of system failures is not dependent
on the availability of any one component.
- Automated control
of data mirrors - The complicated tasks of establishing, suspending,
reversing, and resynchronizing data mirrors are automatic, thus reducing
the chance of data loss or corruption due to user error.
- Easier execution
of planned system outages - Tools for user-controlled operations
help you gracefully bring individual system components offline for scheduled
maintenance while minimizing the downtime experienced by the end users.
If a catastrophe
hits your data center, HACMP/XD will work in concert with ESS Copy Services
PPRC to automatically recover your business-critical applications and
data. In just minutes your users can resume normal business operations
at your mirrored site. Each component plays the following role in your
Storage Server (ESS) - IBM's enterprise-level storage system provides
outstanding performance while keeping your data available via redundant
hardware, mirrored write caches, and RAID-5 and RAID-10 protection for
- Peer to Peer
Remote Copy (PPRC) - IBM's hardware-based data replication solution
provides real-time mirroring between two or more ESS systems, separated
by distances up to 103 km utilizing ESCON links.
- High Availability
Cluster Multi-Processing eXtended Distance for AIX (HACMP/XD) -IBM's
clustering software provides automatic fallover and recovery within
and between IBM pSeries servers to recover from failures of adapters,
applications, individual servers or even an entire data center.
This paper provides
a step-by-step example of how to use HACMP/XD and ESS Copy Services PPRC
to implement a solution over a metro-wide area to keep an example application
and its data highly available. The configuration consists of three pSeries
Servers and two ESS systems located at separate sites.
This paper is written
for Information Technology professionals who have some experience with
the IBM ESS, PPRC, and HACMP. For more information on these technologies,
please see the sources listed in the References section.