Understand the architecture of the system, the company service bus scene and solution, Part 3
ESB scenario solution
Level: Junior July 2004
Part 3 of this series introduces scenes and solutions to realize Enterprise Service Bus, ESBs, and analyzed this author.
Part 2 Overview Multiple scenarios possible solutions.
The bus work role described in Part 1 provides the basis of these scenes.
Let's continue to build a service-oriented architecture, SOA's business service bus, now let's take a look at the various obvious solution patterns of the scene described in Part 2 (see the reference) .
Each portion of the following describes a solution mode of an ESB implementation. In addition to the Basic Adaptors mode, other simple points (P2P) solutions are simple. Each mode has put forward different use of existing technology, and it also makes considerations in two aspects and transplantation.
Please note that the icon of each solution mode is considered to be separated by the service client and the service providing service. Of course, in many cases, the same system or application can either a service client or a service provider. The illustration is not to rule out the system as a separate client and the provider, but acknowledged that the same system can have two different working roles in different interoperabilities. This distinction is usually important when the decision system is selected, confirmed, and calling services as a client role or as a provider role to receive, process, and respond to service requests.
This part of the solution model is:
Basic Adaptors Services Gateway Web Services Compatible Agent's Enterprise Application Integration Infrastructure (Service Choreograph "Service Orchestration Foundation Full SOA INFRASTRUCTURE
Basic Adapter Solution Mode This solution is integrated through a package or adapter technology to achieve a simple point-to-point (P2P) service, not true ESB. This technique is integrated through WSDL-defined SOAP access or other interoperable product technology (such as IBM WebSphere® MQ (MQ)). If these technologies do not provide local models for service interface (such as WSDL), you will need to use custom models to implement SOA specifications.
Although the design is relatively simple, the benefits that can be obtained from this mode cannot be underestimated. For example, direct integration of MQ or SOAP / HTTP can still be loosely coupled, especially interoperability, is a declaration using an interface. At some point in the future, we can interrupt integration through it for an ESB infrastructure that supports initial integration technology. Control levels can also be implemented on the process level of service naming and addressing.
There are already a wide variety of adapters available, and new adapters can also be created by developing tools or runtime techniques. And allow it to provide support for web service specifications and enterprise application integration (EAI) middleware. It can also be provided to a variety of different types of systems, including the latest distributed application server (J2EE server (such as WebSphere), or Microsoft's .NET system), corporate legacy system (such as CICS?) And Commercial Off-the-shelf Software packages (such as SAP or Siebel). Figure 1 illustrates a general basic adapter solution that includes a new integration using an existing HTTP and EAI middleware infrastructure. Although this figure depicts an internal integrated scenario, if you use HTTP as a communication protocol, or use some Internet-compatible EAI technology (such as MQ Internet Pass-Thru), then the solution can also be applied to external scenes.
Figure 1. Basic adapter solution model describes some aspects of the existing HTTP and unmodified EAI infrastructure as support service bus functions
The implementation technique for selecting the basic adapter The following is some of the options for implementing the basic adapter:
Use the SOAP or EAI functionality directly provided by legacy systems or applications. For example, IBM CICS currently provides support for SOAP, and many systems and application packs support MQ or SOAP interfaces. If the application used to provide access is the application you develop by the user and runs on the application server environment, or as long as the application server runtime environment and application development environment can be used to add a package to the application. For example, WebSphere Studio Application Developer can be used to add XML, SOAP, or MQ support to J2EE applications deployed in WebSphere Application Server (Application Server). If this support is not available or unaware (for example, if the XML conversion is not suitable for processing resources on the existing platform), other architectural layers may be required, as shown in FIG. This may be the application server layer that hosts an adapter integrated with an application or legacy system. For example, Application Developer Integration Edition provides a Java 2 Connector Architecture (JFA) connector technology to access legacy systems (such as CICS), and provide J2EE and Web service interfaces through the WebSphere runtime environment. Figure 2. Other architectural layers performing XML conversion processing If you use a development tool to create your own encapsulars, you can enhance the tools provided by tools: perform universal tasks by creating a frame or a set of utility classes, such as security , Log records, etc. However, this method may cause scope creep, and eventually lead to actually become a service gateway or web service compatible agent developed by the user. When defining the feature of the framework, you need to pay attention to whether the cost of verifying the development and maintenance is appropriate, and the converted to more complex mode is more inappropriate.
See Resources for more information on implementing this mode.
The basic adapter profile is from the front, and this solution mode is the lowest demand for new basic architectures or does not need, and it is all widely supported specifications and techniques. From the reaction, it is handed over to the implementation of the application or a single package.
Since this mode is based on the use of collaborative operational techniques and open standards, it is relatively simple to transplant this mode to a more complex architecture. Mode replacement If the above cannot meet the integrated demand, or some additional functions or service quality needs, the package can not meet the needs. If so, it is logically said that the next step should be a service gateway. If you need a higher ESB function, the Web service compatible proxy or Eai Infrastructure for SOA mode is more suitable.
Service Gateway Solution Mode This model represents a basic ESB implementation, close to the "lowest function ESB implementation" described in Part 1. The service gateway typically supports client connections through SOAP / HTTP, MQ, Java Message Services, etc., but can also be provided by WebSphere Business Integration Adaptors, WBIAs. Supports more integrated integration. The gateway component is a role of the central control point for service route, protocol conversion, and security.
The gateway can be used to provide a consistent service namespace to the client (for example, providing a namespace with the URL of the SOAP / HTTP service), and can provide authorization models to the service. In fact, these services are more than a completely different system. The protocol is provided. These functions provided by the gateway are a significant demand when they need to open services to external partners (such as clients or suppliers). However, when you need to simplify the access from the application to the functionality implemented by multiple systems and technology, these features are also useful within a single business.
A critical gateway function is to convert the client supported service protocol to the service protocol supported by the provider. These protocols may include SOAP / HTTP, MQ or SOAP / JMS, JCA, RMI / IIOP, and the like. The ability of candidate implementation technologies requires evaluation for the required clients and providing protocols.
Figure 3 depicts the service gateway solution mode
Figure 3. Implement ESB using the service gateway mode
The implementation of the service gateway implementation technology service gateway solution mode can be implemented in the following ways:
Use the packaged gateway technology, such as WebSphere Application Server Network Deployment, or WebSphere Business Integration Connection. Web Services Gateway. Many gateway technologies support certain forms of middleware filters or processor programming models to implement custom enhancements. Web Services Gateway provides some configurable middleware features. It also supports the XML-based remote procedure to call the Web service request / response handler defined in the Java APIS (Java Apis for XML-Based Remote Procedure Call, JAX-RPC) specification. Use the application development and the runtime feature of the latest application server technology to implement custom gateways. This may include the same type of adapter as described in the previous basic adapter solution mode. If you need a higher feature, you should consider more complex advanced EAI middleware, such as WebSphere Business Integration Message Broker. Many implementations of this mode are existing in legacy techniques, which typically do not use Web service technology. For example, many organizations construct a router transaction, which provides a simple interface for multiple legacy transactions with a text-like data model. This system uses a custom data format with some XML portable advantages, thereby effectively implementing gateway mode.
Word Word Words From the front, although some gateway technologies must be deployed in a properly elastic manner, this solution is still capable of containing the lowest functional infrastructure. The emphasis on interoperability protocols and open standards also makes the aspects involved in the infrastructure are simplified. Most gateway technologies with many other interface types (such as RMI / IIOP, and JCA), most of the ability to synergy, should also be able to reduce the deployment of other connectivity technology. However, gateway technology often limits a simple one-to-one-image service process for request / response and publish / subscription services. More complex and advanced features, such as message conversion, message dependencies, messages, etc. may exceed the functionality that the gateway technology can provide, or need to perform development in gateway technology in custom scenes.
Most ESB technology recognizes gateway mode and its related functions. With this, the interoperability protocols and open standards are used, and the problem of simplifying the gateway function is not too difficult to portally, the problem of porting to the higher ESB infrastructure.
The replacement mode of the service gateway is the most obvious replacement mode is a web service compatibility agent or Eai Infrastructure for SOA. These modes are more suitable when the demand exceeds the functionality that the gateway can provide, or when the packaged gateway technology range is exceeded. On the other hand, if the actual service involves very small, the simple basic adapter solution may be appropriate.
Web Services Compatible Agent Solution Mode This solution represents advanced complex ESB implementation, which provides all functions of the EAI solution and uses an open standard model. Define what level of EAI functions need to be used through a clear demand for specific occasions to determine which EAI technology suitable for use. Figure 4 illustrates an ESB implementation that is compatible with a Web service.
Figure 4. Esb implementation of features that use web services compatible proxy
Selecting a Web Services Compatible Agent implementation technology Web services Compatible proxy can use implemented techniques as follows:
The implementation technology that is most likely to be used in such solutions is an EAI middleware supported by the appropriate web service, such as WebSphere Business Integration Server. If the web service support is mainly required for external integration, the proprietary features of the EAI middleware can be used internal and combined with the use of the service gateway components to add web service support.
See Resources for more information on implementing this mode.
The advantage of web service compatibility proxy analysis is that there is a wealth of functions in an open standard model. However, although EAI middleware technology is mature, the solution supports open standards, especially more advanced web service standards, such as Web Service Policy (WS-Transaction), WS-Transaction, is not mature. . Therefore, the most important disadvantage of this scene is only to be simply applicable to all situations.
Replacement Mode for Web Services Compatibility If you cannot provide appropriate web service support, the service bus (Service Bus) can be implemented with a more proprietary or custom-made manner by Eai Infrastructure for SOA mode, perhaps to be related to service gateway components. Combine to add a web service interface. In addition, if an open standard is the most critical demand, and some EAI functions (such as conversion, and aggregation) can be completed elsewhere (perhaps in an application or adapter), the service gateway mode may be more appropriate.
EAI Infrastructure for SOA is not always suitable for the use of web service specifications for the reasons discussed in this article. But SOA principles can still be applied to various solutions, which can be either proprietary or customized, or an open standard.
An obvious method has been confirmed by many successful implementation. That is to use EAI technology (but do not exclude other techniques) and combine XML to build a custom SOA infrastructure. As long as the service interface has been clearly defined and there is a suitable particle size, the EAI middleware ensures that the principle of interoperability and location-independent of SOA are met. This method is also significant, because all the function and performance of mature EAI technologies are applied to the flexibility of SOA. This advantage can be applied to SOA to implement new and firm and stable infrastructure, or applications that implement the SOA principles on existing infrastructure.
ESBs implemented with this method can use these important openings, and are also in fact standards and are benefited from them. In fact, this is indeed a way to extensive these standards to existing IT infrastructure, and provide a certain basis for future development:
Many EAI technologies have a wide range of applications, especially in separate tissues, as the EAI technology and open standards have the advantages of interoperability. If appropriate, XML data and message format can be used to help implement interoperability and platform-independent, just like XML helps these advantages in Web service specifications. EAI will be very likely to support some WEB services, so open standard interfaces can be provided in a suitable occasion, especially for the use of the Document / Literal SOAP model to disclose the XML format used. In addition, this access can be provided by adding a service gateway to the solution. Sometimes, you can use Java platform independence to provide client API packages, not only for J2EE environments, but also for separate Java environments, support Java database environments and many other occasions. EAI middleware can support other open standards (such as Java Messaging Service), which may not be widely applicable as web services, but there are still many technical supports.
This method is an important step in the development of the fully open standard SOA infrastructure. Although in a sense, at least consider porting it to the Web service standard, this method is an important step in the development of the fully open standard SOA infrastructure. The transition of EAI and XML technology uses at least a method of processing problems (such as interface particle size, public data model and format, etc.), all of which are important steps in development.
Finally, the benefits of this method will again emphasize. Mature EAI technology provides extremely rich ESB functions through the characteristics of proven performance, availability, and scalability, etc. (process and data model, conversion, content-based routes, service aggregation, and arrangement, etc.). Since these functions are the most important needs, they do not use EAI technology to implement ESB without web service technologies, which is fully compliant from the core of the solution, especially if you need to use a web service. Add Web service support.
Figure 5 illustrates the components included in this solution mode.
Figure 5. Enriched service bus with EAI middleware
Selecting the implementation of the EAI middleware mode The selection of EAI middleware depends on the features of ESB functionality required for a particular situation and a variety of different EAI products (such as the WebSphere Business Integration family).
The key to design activities is that the service interface definition model. In order to follow the SOA principle, a direct interface should be used to define services. Although some EAI technology may provide such a model, in other cases, custom solutions are required. In practice, this is often implemented by using XML mode and combining service identity confirmation, addressing, and business data. However, there are also text solutions that do not use XML solutions, such as legacy systems for some service gateway modes. The function of interface model-independent interface is used to declare how the service request and response should be scheduled to use the EAI infrastructure. Applications require some mechanism to interpret interface definitions and appropriate call EAI infrastructure. Also, these mechanisms can be provided by EAI technology, and optionally, the technique of design, including design and development specifications, or the use of the frame API.
The development and maintenance of the framework API obviously costs, but it is more effective than implementing specifications across multiple applications. Such methods are most effective if at least most applications that connect to the service bus supports the same programming language (such as Java).
The use of business data models also needs to be selected, which is based on XML-based, proprietary or customized. One of these models may be beneficial because there are many ordinary and specific industries XML data models. However, many of this model is in the process of transplanting the web service specification, if considering the use of this solution mode is due to the available web service technology, it is not suitable for some reason, then it is not possible to select those specification. Finally, if the web service or other specification-based certain formal access requires the use of this custom mode implementation, you can choose the web service support provided by EAI technology, or add an explicit service gateway component ( If it can better match the demand).
EAI Middleware Mode Analysis Since this solution model represents important development, implementation and maintenance, it is necessary to consider it. The advantage of this mode is that it is completely consistent with the SOA principles, which is repeatedly proved to be beneficial to delivery, and can be implemented with mature technology, enabling it to enable enterprise-class features, elasticity and performance.
There are two main aspects of the cost of the solution. First, the initial implementation of the solution and subsequent maintenance, followed, in the transplantation, with the maturity of web service technology and increasing attention, it is ultimately likely to adopt an open standard solution.
This mode is an instant decision, which depends on whether the recent or medium-term interests prove to be worthwise investment. How much investment depends on the existing level of the EAI used, and it depends on how much the amount of work is added. Recent or mid-term definitions rely on when individual organizations think that emerging Web service specification is enough to meet their functionality or non-functional needs.
The replacement mode of the EAI middleware Web service compatible proxy mode is an implementation of using open standard technology similar to the EAI middleware mode.
Service Orchestration Solution Mode The mode consists of a dedicated service arranging component. This component is not a real ESB, but it supports connection of services through a variety of protocols such as SOAP / HTTP or MQ, which requires or implies the existence of ESB. In some scenarios, this support is sufficient to connect directly to the service provider and server request. However, if this is not the case, the ESB can be provided by any other solution mode described herein. This constitutes a complete SOA infrastructure solution mode.
Figure 6 illustrates the implementation of service arrangements.
Figure 6. Implementation of service arranging
The most important choice to do in this solution mode is that the most important choice to do in this solution mode is that it needs to open standard levels. There are three scenarios:
Web service specification is used to model the service interface and process modeling. The service interface uses a web service specification and combines the use of proprietary process modeling techniques. Use proprietary interfaces and process modeling techniques. Because the Web Service Standard related to the process modeling (first is the Web Services Execution Language for Web Services, BPEL4WS) is recently, to support their products is not mature enough, so these issues are Solution mode is especially related. Most service-proof providers will provide dedicated and standard-based hybrid technology. For example, such techniques include:
WebSphere Enterprise Process Choreographer technology provides support for Web service interfaces and process definitions. MQ Workflow provides support for more mature but more proprietary service, which has both Web service interfaces and proprietary interfaces.
If a proprietary technology is used, it may add a service gateway component to provide web service connectivity to address scalability or elastic demand. If you select a service-edging technology that cannot provide sufficient integration for the service provider (eg, legacy system or application server), you need an ESB in one of the other solutions.
Service Arrange Catalog This solution model relies greatly dependent on whether or not the standard or proprietary solution is implemented. Code-based solutions are currently less mature, but it will eventually be able to provide better interoperability. The proprietary solution will provide scalability and elasticity in a more familiar model, as well as high interoperability communication technologies (such as MQ). However, with the maturity and spread of open standard technologies, this program may eventually need some transplantation.
Complete SOA Infrastructure Solution Mode This model represents the combination of service arranging components and service bus implementations. Because these two aspects have been described in this article, there is not more description here, but it is obviously possible to say a complete implementation. One end uses a fully proprietary solution, which uses EAI Infrastructure for SOA mode of ESB and proprietary service, and other ends of the fully open standard solution, which uses ESB and WEB services for service-proof technology that adapt to open standards. mode.
The main stages of SOA and ESBs are described herein to construct ESB from instant demand, which is just a first step in achieving more complex SOA. This part discusses some useful options for organizing how to develop in a controlled and gradual way. I don't recommend that all organizations use a route. Instead, I want to discuss some questions that should be considered in design SOA or ESB routes.
It is determined that the direct scope involved, achieving a comprehensive SOA mainly presents two aspects: full function, flexible infrastructure implementation and all related functions are disclosed in the form of service. Although it is not completely independent, there is a certain degree of loose coupling between two aspects, which enables the organization to be more flexible to choose how to achieve them.
In some respects, the first decision is to verify the ESB technology or to verify the SOA principles of the functional architecture? This problem introduces two extreme methods:
Rich infrastructure, limited function - herein, mainly involved in verifying these technologies. The infrastructure is likely to include complex ESB functions (open standards or proprietary) and may constitute a full SOA infrastructure solution mode. However, this technique is highly risky and may contain relatively immature techniques. Therefore, do not use it to achieve critical business projects or features. The function is disclosed in the form of a service that has both low hazards and mainly passed through optional pathways. With the verification and maturity of the infrastructure function, the service will be ported to the infrastructure. Basic infrastructure, rich features - Here, mainly involving business functions in the form of service, so you can use new ways to access or combine these functions to deliver business value. In this case, the estimated business benefits or other factors driven changes tends to be very important to reduce technological risks. Therefore, the implementation of the infrastructure or only the most basic most mature web service specification, or uses a more certain EAI technology. Once the infrastructure is appropriately and supports service interoperability, its function can be upgraded or transplanted in the future with the ripening of the ESB technology. Of course, there are two other extreme ways - nothing, or immediately do everything, but these may not be interested in the point of view.
Another method is used intangible. That is a single department, project or engineering progressively adopt SOA and ESB principles, technology and infrastructure. Many organizations may use in the ESB or SOA process, which may be used instead of direct use. This part of this topical technology or practice can provide a more successful test, which is much better than the big-bang method. This is similar to a rich architecture, limited functionality, but it consists of many basic architectures instead of a single rich architecture.
There are two aspects of implementing SOA to determine early: Provide internal or external access to the service, and a method of solving service particle size.
A decision to support internal or external access will drive some factors, including what level of service security (see impact ESB security issues), and whether explicit service gateway components are needed to control external access. As discussed in EAI Infrastructure for SOA Solutions, external access also drivers usage of web service specifications. However, internal access may be more flexible, such as MQ, RMI / IIOP, or proprietary XML.
The problem of service granularity has been widely discussed in the industry (see Referring). Synthetic SOA may contain a variety of particle size services, from technical operability, such as login, accounting, etc.), through business functions (such as query account settlement), to business processing (such as handling stock order).
Each granular level service consists of a lower granular level service or other function. Therefore, there is a need to consider some of the different grades of service aggregation or arrangement, which may be suitable for more than one implementation technology. One utility that can be handled in any particular case is that confirmation, characterization, and naming the available different service granular levels. Then you can define aggregation or proofing requirements between two different granular levels, and choose the appropriate implementation technique.
At the end of this section, it is worth noting that the service implemented from the top-down and the bottom-up model. Self-processes focus on the functionality of applications and legacy systems in the form of service. In general, this involves the use of adapters and development environments to provide appropriate interfaces and leads to the enabled relative fine granular service function.
Since the top and bottom method more involves parsing the business system and components to confirm the process and service architecture processing. This tend to determine the coarse particle size, and these coarse granular services may be composed of more fine granular services. Many organizations may use both methods to perform service confirmation and enable, some intermediate competency points need to be merged. If this aggregation point is explicitly confirmed and classified for a variety of particle size levels in the agreed place, it may be more simple to technicians.
The important phase of the SOA must have a lot of phases in a number of phases to achieve intact SOA. This part points out and discusses some important phases. Because they are largely independent, they do not sort them in the order of the factors that affect individual organizations.
Standard security model - Although simplified or proprietary models may meet the needs in the short term, comprehensive SOA must have a comprehensive and open standard security model. It is a key part of the entire implementation plan to understand which products that support Web service security specifications can meet the needs of the organization. Enable service legacy systems and applications - The development of modern application servers (for example, J2EE servers, such as Application Server) allows organizations to access the database using SQL and JDBC or ODBC interfaces. Similarly, the development of SOA will drive organizations to serve the legacy transactions and application functions. So organizations can plan to define and implement the most appropriate form of service to enable each system. You can choose to use an adapter (such as a JCA adapter) to provide legacy system connectivity using an adapter (such as a JCA adapter) or using EAI gateway technology. To achieve high quality service infrastructure - So far, the most mature web service support available for unreliable communication protocols (SOAP / HTTP specifications provide higher quality services, such as Web services reliable messaging (WS-ReliableMessaging) The Web Service Transaction (WS-Transaction) has no extensive support. EAI technology needs to be used to provide higher quality services for SOA. In the long run, organizations should maintain parallel development in the support of emerging specifications and the strengthening of EAI technology that meets Web service specifications. Determined service granularity level - As determined above, it is very important to determine the particle size level of SOA and aggregation and arranging between the levels of the SOA. The implementation and related processes of each granular level (for example, technical functions, business functions, business processes, etc.) have also constituted a key milestone.
SOA implementation steps The following issues have been discussed in both sections, and you can now constitute a general route implemented by the SOA and service bus:
Deciding which elements of the SOA technology or SOA function should be prioritized (see Determining the direct range involved). Specify or define a suitable item to implement the first solution, which can be both technical tests, business tests, or a real business project that can accept risks. Specifies that ESB scene in SOA can be used for project. More about driving ESB architecture and design decisions and demand analysis of ESB function models. Then select a solution based on these analyzes. Based on more analysis and security and non-functional requirements (see Security issues affecting ESB), you will finally choose a suitable implementation technology. Parallel with this work is to start regulating a route from the first SOA development that implements complete synthesis. This work depends on the focus of the initial test, and contains multiple aspects of the technical function of the infrastructure, or other functional services to utilize the initial test. In both cases, the route should be included in the SOA important phase indicated above.
In addition to the initial project range, it is also planned to develop in several other directions:
Develop and improve data models and processes across organizations. Implement the application phase service and incorporate it in the infrastructure. Develop the technical capabilities of the SOA infrastructure. This series ends here. In this series, how to best use ESB from the viewpoint of the architecture. In the future, with the development of the technology after SOA, you will see more research in the use and solution of ESB.
Get the product used in this article If you are a subscriberWorks subscriber, you can have a single user license and have the right to use WebSphere Studio Application Developer, WebSphere Application Server, WebSphere WebSphere Business Integration and DB2?, Lotus?, Rational Tivoli? And WebSphere Products - including Eclipse-based WebSphere Studio IDE - to develop, test, evaluate, and verify your application. If you are not a subscriber, you can subscribe now.
Reference
To learn more about the enterprise service bus, see "Understanding the Architecture of the Services" Enterprise Services Bus Scene and Solutions, Part 1 "(DEVELOPERWORKS, June 2004). Read the "Understanding the System Services Engineering Series and Solutions, Part 2" (DEVELOPERWORKS, June 2004) to continue to study ordinary observed scenarios and solution models, and analyze these observations. To understand how Web service technology provides the necessary basis for the implementation of service-oriented architecture, please read the Web Service-Oriented Architecture - The Best Solution to Business Integration of Annrai Otoole - The Best Solution To Business Integration. In the CBDI Forum, please refer to Lawrence Wilkes. SOA - Save Our assets (requires subscription). Patterns: Service-Oriented Architecture and Web Services Red Book, SG24-6303-00, April 2004, author endrei M. et al. Read "Migrate to the Service-Oriented Architecture, Part 1" (DevelWorks, December 2003) and "Migrate to Service-oriented Architecture, Part 2" (DeveloperWorks, December 2003). Read the Web Services Description Language (WSDL) 1.1 specification. Check the Business Process Execution Language for Web Services Version 1.1 specification. Access to LooselyCoupled.com to get a list of Enterprise Service Bus (Enterprise Service Bus). Initially defined the Enterprise Service Bus (ESB) on Gartner, you need to subscribe, but you can also find their site, the title of this article is Predicts 2003: Enterprise Service Buss Emerge, 2002 December 9th released, author ROY W. SCHULTE. Study the IBM Patterns for E-Business website. Access IBM's on-demand Business (On Demand Business) for more information on the on-demand operating environment. The information used herein is from Forthcoming Red Books Patterns: Implementing An SOA Using An Enterprise Service Bus, SG24-6346-00, draft, author Martin Keen et al. Using Web Services for Business Integration Red Book, SG24-6583-00, April 2004, Author Gert Van de Putte, etc. WebSphere Version 5.1 Application Developer 5.1.1 Web Services Handbook Red Book, SG246891-01, February 2004, Author UELI WAHLI, etc.
Check the articles written by Jeff Hanson on the Builder.com website Coarse-grained Interfaces Enable Service Composition in SOA. Learn more about Enterprise Integration Solutions on the IBM website. For more information on web services, read "An Introduction to Web Services Gateway" (DeveloperWorks, May 2002). See the WS-ReliableMessaging specification. Check the WS-Transaction Standard. Reading specification WS-Policy. See the WS-Security specification. Find other SOA and Web services technology resources in the SOA and Web Services technology area of DeveloperWorks. Visit WebSphere Studio Zone on developerWorks for comprehensive information on how to get and use the IBM application development tool. Purchase Discount books in the Developer Bookstore. About the author Rick Robinson is an IT architect of IBM, and he joined the company in March 1997 as a developer. He is a member of the Architecture Services Group of Emea WebSphere Lab Services. He started paying attention to it when the 1999 WebSphere software platform was first published. Rick is more concerned about technology instead of the industry, but in the past three years, he spent a lot of time and financial fields working together. Rick is a trusted IT architecture design professionals in IBM. Before joining IBM, the Rick was pursuing PhD of Physical.
