IBM BPM, V8.0.1, All platforms > Install IBM BPM > Plan for IBM BPM > Plan the ND environment

Considerations for selecting a topology

Selecting an appropriate topology for your deployment environment depends upon several factors.

When you select a topology pattern, consider the following factors:

The Remote Messaging, Remote Support, and Web four-cluster topology pattern is the preferred topology for IBM BPM Standard and IBM BPM Advanced, but the choice ultimately depends upon your individual requirements. The four cluster topology is also the default for the configureNode command, except for z/OS . If you select a three or four cluster topology, configure a routing server for Business Space.

The IBM-supplied topologies can be applied to both Process Server and Process Center topologies. Therefore, your Process Center and Process Server ND environments can be organized in a similar way. The procedures for creating environments for Process Server and Process Center based on IBM-supplied topologies are also similar. The only difference related to IBM-supplied patterns is the recommended patterns for a production environment, and the components configured on the clusters for those patterns.

For information on the components, features, and functionality available in each of the IBM BPM configurations, see IBM BPM V8.0.1 configuration capabilities


Condensed topology pattern selection criteria

Consider the information listed in the following table, which is a quick guide to selecting your production topology. This table provides a condensed list of the advantages and disadvantages of each of the topology patterns.

For information about which BPM products support the supplied topology patterns, see Topology patterns and supported BPM product features.

Considerations for selecting a topology for your deployment environment
Consideration Topology Pattern
Single cluster Remote Messaging Remote Messaging and Remote Support Remote Messaging, Remote Support, and Web
Number of clusters to maintain

One cluster for all components

Two clusters:

One cluster for applications and for the support infrastructure

One cluster for messaging

Three clusters:

One cluster for applications

One cluster for the support infrastructure

One cluster for messaging

Four clusters:

One cluster for applications

One cluster for web interfaces

One cluster for support infrastructure

One cluster for messaging

Hardware requirements

Can be implemented on limited hardware

More hardware required for distributed environments

More hardware required for distributed environments

Most hardware intensive

Asynchronous interactions

Use should be minimal

Use must be balanced against resource availability

Ideal environment for asynchronous interactions

Ideal environment for asynchronous interactions

Long-running processes, state machines, and human tasks

Use should be minimal

Use must be balanced against resource availability

Ideal environment for interruptible processes, state machines, and human tasks

Ideal environment for interruptible processes, state machines, and human tasks

Heavy CEI activity

Not recommended (Light CEI use should be balanced against resource usage.)

Not recommended (Light CEI use should be balanced against resource usage.)

Ideal environment for heavy CEI use

Ideal environment for heavy CEI use

Administrative burden

Relatively small

Requires additional effort

Requires additional administrative effort

Requires most administrative effort

Scalability

All components scaled at the same rate

Messaging cluster scalability limited (no benefit beyond three servers)

All other components scaled at the same rate

Easy to scale

All functions separated

Messaging cluster scalability still limited (no benefit beyond three servers)

Easiest to scale

All functions separated

Messaging cluster scalability still limited (benefit comes when other BPM products are introduced)

Routing server

Optional for load balancing and high availability

Optional for load balancing and high availability

Required for Process Portal – and can also be used for load balancing and high availability

Required for Process Portal – and can also be used for load balancing and high availability

The routing server can be IBM HTTP Server, WebSphere Application Server proxy server, or any other web server, reverse proxy server, or network layer IP sprayer.

Plan the ND environment


Related concepts:
Topology patterns and supported product features
Load balancing and failover with IBM HTTP Server
Topologies of an ND environment


Related information:
Configure Process Portal