Let’s evaluate the following question: Is an Enterprise Service Bus (ESB) a technical tool implementation that aids in delivering a Service Oriented Architecture (SOA)?
This article looks at ESB and SOA from a conceptual perspective as opposed to practical implementation. Determining the right solution means for these concepts to be fully investigated and applied to the specific enterprise and its landscape. It is therefore by no means the final say on the subject and does not prescribe a specific tool.
In order to evaluate this statement, one needs to consider what ESB and SOA respectively are, given a proper definition of the global concept of Enterprise Application Integration (EAI).
Enterprise Application Integration (EAI) – Useful to clearing up the confusion of integration semantics:
Let’s firstly clarify that Sirius does not consider EAI to be an appraoch like SOA or ESB. We consider it to be the broader concept called Enterprise application integration (EAI) since, by definition, EAI is the use of technologies and services across an enterprise to enable the integration of software applications and hardware systems. EAI can therefore be related to middleware as well as distributed technologies. Despite our definition, many vendors offer ‘EAI suites’ that provide cross-platform, cross-language integration solutions.The sharing of data and business processes between applications are the primary purposes for these solutions while also prescribing a set of principles for integration of multiple systems for communication architectures, such as message-oriented middleware (MOM). In our opinion, this blurs the line between EAI (a concept/framework), which spans various layers of application, business process, data, information and integration architectures, and ESB which is essentially a tool to enablling EAI. Hence, such EAI suites should really be called ESB tools.
Developing EAI technologies involve web service integration, service-oriented architecture, content integration and business process workflow integration. EAI is usually challenged from a technology perspective by different operating systems, database architectures and/or computer languages, as well as other situations where legacy systems are no longer supported by the original manufacturers. Many types of business software such as supply chain management applications, ERP systems, CRM applications for managing customers, business intelligence applications, payroll and human resources systems typically cannot communicate with one another in order to share data or business rules. .
EAI aims to establish the required communication by meeting these challenges through the fulfillment of three purposes, as follows:
- Data Integration: Ensures consistent information across different systems.
- Vendor Independence: Business policies or rules regarding specific business applications do not have to be re-implemented when replaced with different brand applications.
- Common Facade: Developers and users are not required to learn new or different applications because a consistent software application access interface is provided.
The advantages of EAI are clear:
- Enabling real-time information access
- Streamlining processes
- Accessing information more efficiently
- Transferring data and information across multiple platforms
- Simplified development and maintenance
Management strategies to enterprise integration must address the four different domains depicted below. Each of these domains represents different challenges.
Relation between the architecture models, viewpoints and integration distribution models from an Enterprise Architecture (EA) perspective.
Service Oriented Architecture (SOA):
A service-oriented architecture is essentially a collection of services. These services communicate with each other. The communication can involve either simple data passing or it could involve two or more services coordinating some activity. Some means of connecting services to each other is needed. At a more technical level, SOA is an evolution of distributed computing based on the request/reply design paradigm for synchronous and asynchronous applications which utilise these services to communicate and to share information and functions. SOA can be applicable to a single technology (eg. SAP), with its various modules that communicate via prescibed types of services on fixed standards, or it can be a collection across an enterprise for various applications to integrate via services. SOA developed as an approach to allow scalabiality following the early main-frame computing era where integration challenges resulted into hardly supportable ‘spaghetti’ code, performing non-real-time, flat file processing in order to accomplish integration.
Hence, SOA is one approach to enabling EAI. Clearly, different types of frameworks can be developed to enable EAI. In theory, perfect EAI can only be enabled by tightly controlled frameworks. Frameworks imply governance, which is a big factor of SOA, however,usually lacks completely by design from SOA development. There are different definitions for SOA, however, we persist that no single SOA operates with complete guidance for integration unless clearly stipulated and therefore does not suffice under the definition of a fully fledged framework, catering for all integration that might realistically be required to enable fully fledged EAI. We therefore view SOA as a principle approach to enable EAI requirements. Various different integration technologies and integration methods may still be at play to enable SOA. One may argue that Service orientation is based on the software development concept of Object Orientation (OO). Just like OO can enable re-usability and scalability of code, SOA can enable re-usability and scalability of live, compiled services to remove duplication in information flow across the enterprise. However, just like OO cannot guarantee that code gets re-used, similarly SOA cannot guarantee that services get re-used, since SOA’s set of services could potentially all still just be enabling point-to-point integration as opposed to agile integration (plug-and-play). This is where an Enterprise Service Bus (ESB) might come in handy to either compliment or take over from the distributed, standardised services composing your SOA model.
Enterprise Service Bus (ESB):
An Enterprise Service Bus (ESB) is fundamentally an architecture. However, numerous tools have now been developed to practically enable it as a freestanding technology to solve EAI challenges. It is a set of rules and principles for integrating numerous applications together over a bus-like infrastructure which can also house the various services contained in SOA, together with other integration mechanisms in order to centralise and standardise integration across the organisation. Essentially, ESB provides Application Programming Interfaces (APIs) for developers to create services and send messages between services with real-time integration. Various ESB products enable users to build this type of architecture, but vary in the way that they do it and the capabilities that they offer. The core concept of the ESB architecture is that you integrate different applications by putting a communication bus between them and then enable each application to talk to the bus. This decouples systems from each other, allowing them to communicate without dependency on or knowledge of other systems on the bus. The concept of ESB was born out of the need to move away from point-to-point integration as well as improvements to SOA, which can become brittle and hard to manage over time. Point-to-point integration results in custom integration code being spread among applications with no central way to monitor or troubleshoot. This is often referred to as “spaghetti code” and does not scale because it creates tight dependencies between applications. ESB solves this by forcing centralised control for integration testing. Essentially, ESB should be intended to form the backbone of any SOA architecture to enable centralised and standardised control, in turn meeting the challenges and gaining the advantages of properly governed EAI. ESB properly spans the various domains of integration as opposed to only grouping services. We need to place a cautionary note here that no ESB will cater for every aspect of integration across all technologies and at maximum offers control for the majority of integration points within the ringfenced control of an organisation. Some forms of batch processing might still be required internally due to complexity, skill or speed constraints. Let’s discuss the overlap of ESB and ETL for big data briefly in order to eliminate confusion on what this now means in the cloud- and big data context.
![esb[1]](https://i0.wp.com/siriussa.com/wp-content/uploads/2018/02/esb1.jpg)
ESB and SOA in the new Big Data Context: APIs and ESBs can be used to Simplify Data Integration for ETL in the cloud computing domain
The increase in popularity of Application Programming Interfaces (APIs) has made it much easier to create connectivity. With APIs, developers can access endpoints and build connections without having in-depth knowledge of the system itself, simplifying processes tremendously. As Extract-Transform-Load (ETL) data tools remained focused more towards BI and big data solutions, and as traditional operational data integration methods become outdated with the rise in popularity of cloud computing, ESBs become better options to solving connectivity requirements.
An enterprise service bus (ESB) provides API-based connectivity with real-time integration. Unlike traditional ETL tools used for data integration, an ESB isolates applications and databases from one another by providing a middle service layer. This abstraction layer reduces dependencies by decoupling systems and provides flexibility. Developers can utilize pre-built connectors to easily create integrations without extensive knowledge of specific application and database internals, and can very quickly makes changes without fear of the entire integrated system falling apart. Shielded by APIs, applications and databases can be modified and upgraded without unexpected consequences. In comparison to utilizing ETL tools for operational integration, an ESB provides a much more logical and well defined approach to take on such an initiative. On the down-side, batch data transfers via ESB can become painful and ETL developers such as Business Intelligence (BI) developers, who are used to typically accessing large quantities of data directly, may be prone to circumvent the ESB, therefore forcing integration architecture back to only being conceptual and responsive rather than physical and in control. It therefore remains imperative for collaboration between BI and integration team leads, since the most large scale ESB tools now include options to connect to relational databases as well as emerging Big Data platforms. For instance, take a look at the following comparison between Biztalk and Informatica.
In Summary
- SOA is an architectural approach where we expose ‘service’ in a coarse-grained manner whereas ESB is a technical implementation that aids in delivering a SOA.
- SOA brings cost effective, reusable and low lead time solutions to an organization whereas ESB enables low cost integration and benefits companies with limited IT resources.
- SOA is a way of building the next generation of applications from ‘lego blocks’ called services whereas ESB is a piece of infrastructure software that provides APIs for developers to create services and send messages between services.
- SOA is just like a car and ESB is like a road on which this car runs.
- SOA is an architectural model for implementing loosely coupled service based applications whereas ESB is a piece of infrastructure software that helps developers to develop services, and communicate between services through suitable APIs.
- Certain ESB tools are now allowing SOA to be implemented while also enabling big data ETL via the necessary exposed layers for data enquiry
Although one can trace back geometric approaches for data analysis quite far in the past, TDA started with the pioneering works of Edelsbrunner et al. (2002) and Zomorodian and Carlsson (2005) in persistent homology (genetic sciences) and was popularized in a landmark paper in 2009 Carlsson (2009). Carlsson noted that an important feature of modern science and engineering is that data of various kinds is being produced at an unprecedented rate 
An oversimplified way to explain this to the non-mathematical academic community is to say that TDA is like deriving knowledge from the structures underlining a piece of constantly changing origami. If the points on the origami describes the origami itself, then the relationships between those points tell us something about what those points are and why they are where they are. The ‘what’ and the ‘why’ can be seen as knowledge, which we can use to generate new, creative, and meaningful origami pieces or to learn why things are how they are. This’ origami piece’ can present itself in numerous practical problems of knowledge including DNA, chemistry, human interaction and many more. Associated technologies are therefore being developed, specific to analysing and deriving knowledge on the basis of TDA for specific industries. This enables data scientists to create intelligent, non-linear computational models to effectively analyse massive datasets in order to produce well-informed recomendations to business decisionmakers. Essentially, TDA has the ability to completely transform what your actuary used to do from a data-drives-value perspective. It has the ability to allow data science to generate value from the following key aspects that intrinsically differentiate data science from traditional data analysis: Utilizing erratic, sporadic as well as consistent and pattern-based data input to produce learning, usable for well-informed, robust decision-making. This means, for example, that we can also analyse data generated outside of an organisation to find relationships affecting the organisation itself. The ability to analyse seemingly random datasets is a valuable tool when seeking innovative opportunity.![zigbee-internet-of-things[1]](https://i0.wp.com/siriussa.com/wp-content/uploads/2018/02/zigbee-internet-of-things1.png?resize=300%2C164)
The mathematical branch of topology studies only properties of geometric objects which do not depend on the chosen 
If ‘data’ refers to singular items of values without context, then Data Architecture is necessary to contain and organize the various data resources into a manageable data environment.
The Data Architect focuses on the technical depth implied by and associated with the design, physical structuring, storage, movement, planning, backup, retention, dissemination and classification of all data elements within the enterprise and may or may not be required to engage across business functions. Such functions would typically result into technical specifications during business application design.
Recent Comments