A pattern is a reusable solution that encapsulates a top-down tested approach to solving a common architecture, design, or deployment task in a particular context. This approach complements bottom-up development of creating message flows and nodes.
A number of patterns are supplied in the WebSphere Message Broker Toolkit, and you can use the Patterns Explorer, which includes comprehensive help, to simplify creation of common scenarios.
You can configure these patterns with values for use in your own environment to solve specific business problems. The supplied patterns use preferred techniques in message flow design, to produce efficient and reliable flows.
SCA nodes for WebSphere Process Server
Five new built-in message flow nodes are provided to improve the interaction between WebSphere Message Broker and WebSphere Process Server Version 6.2 by using Web Services (SOAP over HTTP) or WebSphere MQ bindings.
The nodes are the SCAInput, SCAReply, SCARequest, SCAAsyncRequest, and SCAAsyncResponse nodes.
Enhanced support for the PHPCompute node
Support for the PHP scripting language is available on all operating systems on which WebSphere Message Broker is supported, except Solaris on x86-64. The PHPCompute node supports general-purpose transformation logic in the PHP language, and complements the Compute, JavaCompute, XSLTransform, and Mapping nodes. In addition, the set of supported PHP extensions has been increased.
SAP Enhancements:
Use the SAPReply node to send a reply to an SAP synchronous callout. Use this node with an SAPInput node to implement a message flow application that acts as a remote function call (RFC) destination.
Generic IDoc routing
By using the SAPInput node in passthrough mode, WebSphere Message Broker can receive any IDoc, and route it according to IDoc type. By using this method, you can also use a single RFC program ID to receive all IDoc types, while still allowing individual IDoc processing.
SAP high availability
You can deploy an SAP adapter and a message flow that contains an SAPInput node to two brokers on your network; these brokers can accept IDocs concurrently from the same SAP system so that you can build a highly available environment.
On distributed systems, two brokers share state by using queues on a third queue manager, which is running in multi-instance mode. Each broker has client connections to that queue manager.
On z/OS®, the shared state is stored on a shared queue. Each broker connects to the queue sharing group.
Iterative discovery
You can take an adapter component that was created by using the Adapter Connection wizard in WebSphere Message Broker Version 7.0, and update it with newly discovered objects from the Enterprise Information System (EIS) by running the Adapter Connection - Iterative Discovery wizard. This facility is known as iterative discovery. You can either add the new objects without modifying existing objects, or replace existing objects.
Iterative deployment
If your message flow acts as a gateway to an EIS, you can use it to call new services that did not exist when you developed the flow. You can also create an event handler to an EIS to handle new event types that did not exist when you first developed your message flow. In both cases, if a new service or event is provided by the EIS, you do not have to modify and retest the message flow. This facility is called iterative deployment.
Sequence and Resequence nodes:
WebSphere Message Broker provides support for adding sequence numbers to messages, and for reordering messages in the message flow based on their sequence number. You can use the new Sequence node to add sequence numbers to the messages, and the new Resequence node to reorder the messages into their original sequential order.
New configurable services for EDA nodes:
You can use the following configurable services to define the WebSphere MQ queues on which EDA nodes store event state:
Aggregation
Collector
Resequence
Timer
Resource statistics:
You can collect statistics for some of the resources that are used by execution groups in the broker to help with problem diagnosis and broker optimization. Supported resources are the Java™ Virtual Machine (JVM), and the outbound sockets. For example, you can monitor the sockets that are used by SOAP nodes in your message flows.
You can start and stop statistics collection at broker or execution group level by using the WebSphere Message Broker Explorer, the CMP API, or the mqsichangeresourcestats command.
The resource statistics framework is based on the existing accounting and statistics for message flows, and generates periodic messages as publications that your programs can subscribe to. You can also view these statistics in the WebSphere Message Broker Explorer, which provides both numeric and graphical representations.
Multi-instance brokers
WebSphere Message Broker builds on the multi-instance queue manager support introduced in WebSphere MQ Version 7.0.1 to provide a highly available configuration with active and passive brokers.
Multi-instance brokers and queue managers store their configurations on shared network storage so that if a failure occurs in an active component, the passive component assumes the configuration and operation of the active component. The use of queue managers in this way avoids the requirement for a high availability solution, such as HACMP™, supplied by a vendor software company.
----
Java
