Ibm Hacmp Aix 5l

Proven availability for an on demand environment

HACMP for AIX 5L Simpler. Faster. Goes the Distance.

Need for high availability Throughout the day – often 24 hours a day – IT systems are hard at work. But what happens when the systems or applications fail? It is widely documented that unplanned outages increase the total cost of IT ownership. In fact, surveys indicate that an outage

Highlights

can cost from several thousand to over a million pounds per hour, and some

■

Helps reduce unplanned

■

Backup systems can be located

businesses virtually shut down if their

outages and improve system

at a remote site for geographic

systems fail. And although hardware

availability

disaster recovery

has become very reliable, research shows that unplanned outages still

■

Offers ease of use through

■

Provides failover on demand to

occur and typically result from operator

configuration wizards, auto-

enable system maintenance

error, software bugs, environmental

discovery and a Web-based

without service interruption.

conditions and so forth – problems that

interface

reliable hardware cannot prevent. Even planned outages for application and system maintenance can impact business performance or delay necessary updates. Service availability is critical to a reliable, cost-effective IT infrastructure. At the same time, clients want solutions that simplify the IT environment and enable the flexibility to respond to changing business needs.

IBM’s premier availability solution

and reverse the mirroring process. This

Business continuity and disaster recovery

IBM High Availability Cluster

minimises expensive downtime for both

with HACMP/XD

Multiprocessing for AIX 5L V5.3

planned and unplanned outages and

HACMP/XD, an optional feature of

(HACMP V5.3) helps protect critical

provides flexibility to accommodate

HACMP, is essential for clients who

business applications from failures. For

changing business needs.

wish to protect themselves from site-

over a decade, HACMP has been

wide failures or large-scale disasters by

providing reliable monitoring, failure

HACMP can be configured to react to

mirroring business-critical data across

detection and automated recovery of

hundreds of system events including

sites and enabling failover to these

business application environments to

problems that are not severe enough to

remote sites. This applies to businesses

backup resources. And the HACMP/XD

interrupt proper system operation (such

of any size with multiple sites or

(Extended Distance) component

as process failure or exhaustion of

regional operations or wherever

extends failover to backup resources at

system resources). HACMP monitors,

decentralisation of data is desired.

remote sites to ensure reliable service

detects and reacts to such conditions,

HACMP/XD offers multiple technologies

delivery even if a disaster disables the

maintaining service availability during

for achieving long distance data

primary site. HACMP/XD provides

random, unexpected software

mirroring, failover and

remote failover for Enterprise Storage

problems. Up to 32 servers can

resynchronisation.

Server (ESS) and SAN Volume

participate in an HACMP cluster – ideal

Controller Metro-Mirror (formerly Peer-

for environments requiring scale-out

to-Peer Remote Copy – PPRC) peers

growth with rock-solid reliability.

and unlimited distance failover for IPconnected peers using proven AIX 5L

HACMP can also virtually eliminate

mirroring technology. When HACMP

planned outages by transferring users,

detects a failure, it can quickly and

applications and data to backup

automatically take action to restore the

systems during scheduled

application, restarting it on a redundant

maintenance. HACMP clusters can be

system if necessary. If the backup

configured to meet complex and varied

system is at a remote site, HACMP can

application availability and recovery

also switch to a mirror (replicated) copy

needs.

of the application data automatically

●

HACMP/XD’s new Geographic Logical Volume Manager (GLVM) provides remote data mirroring and enables failover to remote sites using the mirrored data. GLVM’s IP-based mirroring automatically maintains a data replica across geographically remote sites and completely manages data synchronisation during production, failover and recovery. By integrating with AIX 5L’s native Logical Volume Manager, GLVM minimises the need for separate configuration and management

●

●

●

●

HACMP/XD supports ESS/MetroMirror, enabling automatic failover of disks that are Metro-Mirror pairs and creating a powerful solution for clients on ESS with Metro-Mirror. By automating the management of Metro-Mirror, recovery time is minimised after an outage regardless of whether the clustered environment is local or geographically dispersed. HACMP/XD in combination with Metro-Mirror manages a clustered environment to ensure mirroring of critical data is maintained at all times HACMP/XD also supports IBM TotalStorage PPRC/MetroMirror for SAN Volume Controller which can improve disaster recovery and business continuance capabilities by applying and managing copy services across disparate disk storage devices within the SAN HACMP/XD support of the enterprise Remote Copy Management Facility (eRCMF) for ESS/Metro-Mirror allows users to take advantage of the hardware’s simplified configuration steps through eRCMF and still provides automatic recovery of disk failures in a Metro-Mirror peer environment In addition to the new GLVM, HACMP/XD retains HAGEO for unlimited distance, IP-based mirroring.

High availability cluster enhancements

●

HACMP is tightly integrated with AIX 5L to exploit some of its most powerful features. New features and enhancements make HACMP: SIMPLER to configure and more flexible to maintain:

●

●

●

●

●

●

●

The Two-node Configuration Wizard configures a cluster in four easy steps The Online Planning Worksheet tools walks you through the cluster configuration process and lets you plan and update the corresponding cluster automatically The Cluster Test Tool allows the evaluation of cluster behaviour under a set of specified circumstances Enhanced security mechanisms protect the integrity and operation of a cluster Web-based cluster management enables configuration, monitoring and management of a cluster from a Web browser Automated nightly cluster verification reduces the risk that a change to the cluster may interfere with future cluster operation or failover HACMP File Collections provide an easy way to maintain file synchronisation across a cluster

●

The optional feature HACMP Smart Assist package simplifies the implementation of HACMP in DB2. Oracle and WebSphere environments by reading the application configuration data and configuring HACMP accordingly, providing all the necessary application monitors, start/stop scripts and so forth. All three Smart Assists are included in one inexpensive package Support for Veritas File System to simplify implementation in VGS environments.

FASTER event processing and application recovery:

●

●

●

●

Improved policy-based cluster manager intelligently and automatically responds to failures and removes the need for scriptbased management Up to 40 percent faster failover than HACMP V5.1 in basic clusters, minimising application downtime1 Streamlined resource movement by the cluster manager enables faster response to failures The ability to define cluster-wide dependencies streamlines and simplifies failover of complex, multitier applications.

Goes the Distance to provide disaster

Capacity BackUp for business continuity

Complementary cluster software

recovery across multiple sites:

For selected high-end IBM POWER4

IBM also offers a broad range of

and POWER5 processor-based servers,

additional tools to aid in efficiently

a specially configured and priced

building, managing and expanding

Capacity BackUp system is available for

high-availability clusters in AIX 5L

installation at a remote site. The server

environments. These include:

●

●

AIX 5L Logical Volume Manager split mirroring for disaster recovery in SAN environments The optional HACMP/XD feature provides a portfolio of tools that perform automatic recovery of hardware and software failures across disparate geographic distances. New additions to the portfolio include: – GLVM which provides remote data mirroring and failover – Support for enterprise Remote Copy Management Facility (eRCMF) with ESS/MetroMirror – Support for IBM TotalStorage Metro-Mirror (PPRC) for SAN Volume Controller – New capabilities to manage remote sites and disaster recovery more easily.

contains a minimum number of active processors and a large number of

●

inactive Capacity on Demand (CoD) processors that can be temporarily activated when needed for disaster recovery. HACMP can automatically detect the loss of the production system and activate the Capacity BackUp system to roll client-critical applications over to as many CoD processors as needed to keep the business running with minimal loss of service.

●

Cluster Systems Management (CSM) for AIX 5L provides a powerful tool to manage multiple machines from a single point-ofcontrol. Cluster management capabilities include system configuration, software distribution and maintenance, hardware control, remote command execution, diagnostics and system monitoring with automated problem responses General Parallel File System (GPFS) for AIX 5L is a highperformance, shared-disk file system using standard UNIX® file system interfaces and providing concurrent access to data from all nodes in a cluster. GPFS currently powers several of the world’s most powerful supercomputers and holds the Terabyte Sorting performance record.2 In addition, GPFS can be configured to provide extremely high reliability by eliminating many single points-of-failure (which are typical of network and SAN file systems)

●

●

Workload Manager for AIX 5L provides resource balancing between applications Tivoli Storage Manager provides enterprise management of backup and recovery to tape or disk.

partitioning, outstanding scalability and

Services, an offering which provides

availability, broad support of open

basic and customised assistance for

standards for application flexibility and a

installation of HACMP clusters. This

full range of tools to manage IT

service is customisable with the

infrastructure in an on demand world.

following elements:

Gaining the IBM advantage

●

High Availability Cluster Proof of Concept Review Planning and design of an Availability Cluster Installation and configuration of an Availability Cluster Application integration assistance Development and execution of a Cluster Test Plan Enhanced monitoring and reporting setup Operations planning and operations documentation development Migration/upgrades services.

New generation of on demand servers

High availability solutions from a single

HACMP runs under the AIX 5L

vendor provide clients the confidence

operating system on the IBM _`

that comes from integrated design and

p5, _` i5 and pSeries families –

testing. This reduces the risk of failures

the servers of choice for UNIX operating

resulting from combining disparate

system-based business-critical

components from multiple vendors and

●

applications. These servers offer the

can be a critical factor for business

●

performance, availability, scalability and

environments. IBM high availability

infrastructure management demanded

solutions provide the advantage of

by today’s growing on demand

_` p5, _` i5 and pSeries

business environments. They combine

servers, the AIX 5L operating system,

the benefits of IBM Power Architecture

IBM TotalStorage offerings and

processors and mainframe-inspired

HACMP solutions.

●

with AIX 5L for efficient handling of

The IBM _` family is backed by

Based on an assessment of the

mission-critical applications.

comprehensive offerings and resources

complete system environment,

that provide value at every stage of IT

IBM availability experts can design a

The IBM _` family of servers

implementation. These include High

client solution to meet the target

feature innovative technology,

Availability Cluster Implementation

availability level for on demand

●

●

●

●

RAS (reliability, availability, serviceability)

outstanding performance, logical

business needs.

For more information To learn more about the HACMP for AIX 5L on the IBM _` p5, _` i5 and pSeries server families, contact your IBM marketing IBM United Kingdom Limited emea marketing and publishing services (emaps) Normandy House PO Box 32 Bunnian Place Basingstoke RG21 7EJ United Kingdom

representative or IBM Business Partner or visit the following Web sites:

●

ibm.com/servers/eserver/ pseries/ha

●

ibm.com/servers/eserver/ clusters/software

●

ibm.com/servers/aix

●

ibm.com/iseries/aix

●

ibm.com/common/ssi

The IBM home page can be found at ibm.com IBM, the IBM logo, the e-business logo, _`, AIX 5L, DB2, Enterprise Storage Server, HACMP, POWER4, POWER5, pSeries, Tivoli, TotalStorage and WebSphere are trademarks or registered trademarks of IBM Corporation in the United States, other countries or both. UNIX is a registered trademark of The Open Group in the United States and other countries. Other company, product and service names may be trademarks, or service marks of others. 1

Based on study for IBM by Lakeview Technologies; May, 2004.

2

Details available at: http://www.almaden.ibm.com/ StorageSystems/file_systems/GPFS/ gpfs-spsort.shtml References in this publication to IBM products, programs or services do not imply that IBM intends to make these available in all countries in which IBM operates. Any reference to an IBM product, program or service is not intended to imply that only IBM products, programs or services may be used. Any functionally equivalent product, program or service may be used instead. IBM hardware products are manufactured from new parts, or new and used parts. In some cases, the hardware product may not be new and may have been previously installed. Regardless, IBM warranty terms apply. This publication is for general guidance only. Information is subject to change without notice. Please contact your local IBM sales office or reseller for latest information on IBM products and services. This publication contains non-IBM Internet addresses. IBM is not responsible for information found at these Web sites. IBM does not provide legal, accounting or audit advice or represent or warrant that its products or services ensure compliance with laws. Clients are responsible for compliance with applicable securities laws and regulations, including national laws and regulations. Photographs may show design models.

© Copyright IBM Corporation 2005 All Rights Reserved.

PSD00109-GBEN-03 05-PSE-001079 (KH 08/05)