Data Science, Machine Learning & API / SOA: Insights and Best Practices
Over the years as I have taught, designed and implemented software architecture the question I have been asked over and over again has been:
What is Architecture?
According to the Arsanjani view of Architecture :-), the Architecture of a system is the holistic view of the relative configuration of a set of static and dynamic elements that include what I call the 6C's :
Components -- what are the structural building blocks of a solution, or a style of architecture, of the elements that can be combined to produce a larger structure
Connectors -- The components need to connect with one another, whether statically (as in an Entity-relationship kind of relation) or dynamically, as in a composition in an SOA, where you may have orchestration or choreography.
Constraints -- The constraints on the connectors and/or components that provide rules of engagement of what is permissible and what is not.
Composition -- How to compose or what are the valid compositions of components
Context -- the context of invocation of a component is critical to the designation of how that component will behave
Containers -- the components must live in some runtime container that will provide uniform life-cycle management capabilities for them
Traditionally, the first three have been the main focus in universities and textbooks. But the latter three that I have added seem to be essential to a more complete depiction of the operational or actionable perspective on software architecture.
Let me know what you think.
Ali_Arsanjani 120000D8QB 4,901 Views
To engage in an eco-system, you need to sense the environment, the eco-system. Business sensors gather business significant events and apply business policies to see what actions need to be taken as they sense a threat, an advantage, a change in the business climate, in the eco-system architecture.
Thus, to measure business performance and monitor business agility, you need what I am coining as "business sensors".
Ali_Arsanjani 120000D8QB 7,632 Views
Architectural Paradigms of the Future Part 2: Eco-system Architecture engulfs Enterprise and Business Architecture in the Cloud
Ali_Arsanjani 120000D8QB Tags:  service-oriented-architec... enterprise-architecture business-architecture eco-system-architecture 14,529 Views
In the context of a cloud formation, where is the "enterprise", per se? Of course it exists in the traditional sense. And so does enterprise architecture. But in the context of a cloud, which may span private, public and hybrid instances, where is the "enterprise" now?
Rather than a physical location, the enterprise now seems to be more nebulous and "logical" or "conceptual". It now can be seen to expand possibly, with the cloud, as the amorphous mass expands so does the virtual enterprise. This is where the enterprise merges with the eco-system and you see the emergence and need for addressing aspects of the eco-system that span physical and logical enterprises. As we needed enterprise architecture, for the cloud we need an eco-system architecture.
Eco-systems are more dynamic, opportunistic in some instances; some shorter and some longer enduring alliances between interested business parties. This not only affects the IT side, as it were, in a "bottom up" fashion, but also the business architecture; in a more "top down" manner.
The structure and function of a business is defined by its model of what goods and services it provides and consumes; who it needs to partner with at what point in its dynamic existence. This relates to policies that cut across and senses the eco-system for more appropriate alliances and takes competitive advantage of this new configuration. A business architecture in an eco-system architecture is then configured by rules and policies driven by competitive business goals, not statically defined.
The cloud provides a virtual rockbed for the reconfigurable enterprise that supports its dynamic business architecture to sense and respond, anticipate and adapt to the ebb and flow of trends in the eco-system. The eco-system architecture defines and configures these elements that include the traditional business architecture, multiple enterprise architecture plus a set of rules, policies, that define desirable (to be sought) and undesirable (to be avoided) configurations of the business and IT as they ride on top of the virtual infrastructure provided by the cloud.
This constitutes the eco-system architecture of the future.
As the eco-system of business evolves into a more opportunistic convergence of business partners into an eco-system of value, similar to the notion of a value chain, but a more loosely coupled exchange of goods and services, the need is increasingly being felt for an upfront design of an architecture structure to enable smooth and measurable business operations.
Eco-system architecture defines the foundation of interaction between a set of semi-autonomous business entities that interact with each other using a set of configurable policies that define the valid business interactions among potential and prospective participants, such as business partners: providers, consumers and brokers.
Eco-system architecture extends Enterprise Architecture in a similar way that EA extends the notion of solution architecture. EA is analogous to "city planning" whereas ECSA is analogous to planning how the economy of the whole nation is structured through a set of rules, policies, standards and structures .
Even though SOA has been around for years, the community at large did not have a clear declaration of intent and "values", although individually, there appeared to be a large degree of consensus. Formalizing this consensus, a group of seasoned SOA practitioners got together recently and agreed to a high level declaration of intent.
Service orientation is a paradigm that frames what you do. Service-oriented architecture (SOA) is a type of architecture that results from applying service orientation.
We have been applying service orientation to help organizations consistently deliver sustainable business value, with increased agility and cost effectiveness, in line with changing business needs.
Through our work we have come to prioritize:
That is, while we value the items on the right, we value the items on the left more.
We follow these principles:
© 2009, the above authors, this declaration may be freely copied in any form, but only in its entirety through this notice.
We have just published the SOA Manifesto . This document helps guide the values and principles of service-orientation and service-oriented architecture.
"Service orientation is a paradigm that frames what you do.A set of 6 values are then outlined, followed by a set of 14 principles that help guide the implementation of the values.
When faced with the forces outlines in the value statements, we provide a preference for one rather than the other, even
though both may be viable under certain circumstances.
I will post links to other authors of the manifesto soon.
The IBM Service-oriented Modeling and Architecture(TM) (SOMA) is IBM’s end to end software development and integration method (aka, process life-cycle) based on service-oriented software engineering principles to produce enterprise and individual project-level SOA or non-SOA solutions. This method includes prescriptive guidance applicable during the phases and iterations required in a software development process life-cycle. This guidance includes the activities and tasks to be performed by various designated roles within the life-cycle requiring input work products and producing or updating output work products and deliverables at specified junctures within the life-cycle. The tasks leverage a set of techniques (“capability patterns”) that prescribe and promote the use of best-practices included in the method, in conducting and accomplishing the task and to instantiate the related work products and deliverables.
SOA enables business agility. It enables flexibility of IT systems that support the business architecture. As the business changes, the IT systems required to support the business in a volatile environment of competitive engagement, are less prone to change.
The history of software engineering started with the separation of functionality or processing and data. The data of the processes that manipulate the data were traditionally separate. There came a time when as the pendulum swung emphasis was placed more on the data in one era ended and the pendulum would swing and emphasis would be placed more on process. Eventually object oriented programming, which evolved into object-oriented design and object oriented analysis, played a key role in the unification of process and data. David Parnas was the first to suggest the notion of information hiding. The notion of information hiding was that access to information or access to data was strictly done through the functionality provided by an object. Those that object encapsulated its data and protected it from direct manipulation. It offered a set of operations that you could perform on the data, but you could only invoke those operations do not access the data directly.
Objects often reflected real-world objects. The identity of the real world object was used as an abstraction and implemented in a software system as a software object. Identity often corresponded to a real-world entity although helper objects evolve from the IT constructs necessary to implement the real-world construct in a computer system.
One of the most important principles of object oriented programming was based on a separation of concerns not only in the domain of abstraction but also in terms of separation of interface from implementation. Not only did we separate data from the operations that manipulated the data all in one object, but we did so one step further by separating the interface of the behavior from the actual implementation of that behavior. The behavior or rather its implementation, actually change the state of the object or as it were manipulated the data directly. However, there must be several ways in which we can implement this data manipulation. Even though there may be several ways in which we can implement this manipulation, there is typically one way to represent the interface or the externally visible signature that would be used by a consumer to request a change to be made to the data that the object owned.
This principle is called programming to interfaces rather than implementation.
These interfaces coupled with the notion of composing a set of similar objects that often had interdependencies among them into the next level of encapsulation which was called the component lead to a whole new era of -based development. It exported interface and could contain multiple objects all of which would typically be expected to be related to one another in some logically cohesive fashion. These objects were expected to be highly collaborative with one another and thus made sense to be called located within the packaging of a single component. This allowed not only manageable functionality but also decrease the risk that nonfunctional requirements would be violated.
The promise of object orientation
What is Cloud Computing? Similar to the Initial era in which SOA was trying to be defined, we are now in the era of more definitions. That is not in itself a negative thing, but a necessary part of the formation of new technology as they become mainstream.
Cloud computing is essentially an extension of SOA and service oriented computing. It provides concentrated infrastructure, platform and data center like capabilities offered at a set of services over the Internet. The cloud has several unique characteristics.
The cloud itself needs to be viewed from both a consumer and a provider perspective.
from the perspective of the consumer they are looking for a service-based delivery model over the web in which a highly reliable efficient self optimizing and feed your head up infrastructure capability are made available to them.
From a Provider perspective, the cloud involves two key aspects of self optimization and virtualization using a service-based delivery model.
The service-based delivery model can operate under one of the following kinds of services: platform as a service, software as a service an infrastructure as a service.
in order to connect the consumer with the provider, the provider must publish in some metadata standard information interchange format the quality of service that it can provide and the services that it can provide, so the service consumer can select among various provider of various services they provide and the various qualities of service that are offered for each type of service.
Could computing is fundamentally a service delivery model. It is instantiated by an architecture that provides the services and is consumed by a Service Consumer's own internal systems.
cloud computing is a
service delivery model
in which a virtualized set of
are offered by Cloud Service Providers with the choice of quality of service.
Service Consumers then have a choice
as to what provider to select
Further more, the model should support a seamless portable view of service such that services
from multiple providers can be choreographed by a consumer or broker without having to
know about the inner workings of the underlying cloud infrastructures.
What do you think?
Ali_Arsanjani 120000D8QB 5,211 Views
Amidst the current economic downturn, and in the mist of the survival instinct that is perculating around the world, it is tough to sit back and observe trends.But one unmistakable trend is that countries are shifting from an industry-focused economy (e.g., manufacturing, see the collapse of manufacturing toward a service economy. The impetus for this gradual change include globalization, technological change,and an overwhelming demand for (cheaper and cheaper) services. In observing this shift, it becomes increasingly clear that services and the service economy play an important role in today's and tomorrow's businesses. As a reflection of this trend, a realization or crystalization of the trend, service ecosystems have emerged: including GrandCentral, eBay, Google Base, Amazon.com, SalesForce.com, and SAP Business by Design to name a few. Such marketplaces enable trading products and also services between various legal entities and with consumers (sometimes referred to as the anonymous legal entity). One major challenge for service ecosystems is the fact that services are not products or goods. This has the potential to change the game.
Ali_Arsanjani 120000D8QB Tags:  service-orientedcomputing maturity pervasive services 5,964 Views
SOA continues to evolve through economic downturns and bends in the road. SOA has evolved into a set of underlying principles, concepts, technologies, paradigm and patterns of designing and building software; all predicated on a set of loosely coupled services that collectively support an organization's business processes and goals.These services provides the enterprise with an adaptive or agile capability to morph and transform itself in the face of environmental change.
As we look at the SOA Maturity Model we have called SIMM (Service Integration Maturity Model), we see a clear evolution path:
1. Simple Services (level 4.0 to 4.9) from 2005 -20072. Composite Services (level 5.0 to 5.9) from 2007 - 2009
now we see a lot of interest in paradigms and technologies that are BASED on Service principles, namely
3. SIMM level 6, Virtualized Services, in the form of Software-as-a-service (Saas) and Cloud Computing.
Services have become Pervasive, they can be found in many of the recent initiatives including Mashups, BPM, Infrastructure, Enterprise Architecture, Governance, etc.
The notion of an SOA is finding new applications in new areas and continues to morph and integrate into new things, ingesting new use-cases but retaining the base principles of service-oriented computing.
Ali_Arsanjani 120000D8QB Tags:  soa manners externalization voad variations voa 2 Comments 6,918 Views
In an effort to be responsive to a number of requests for an update to my old manuscript for "Principles of Advanced Software Engineering: Variation-oriented Analysis, Design and Implementation", let me give you folks some background on the subject. [Background]Originally, I wrote a manuscript back in 1999-2000 and used it to teach my classes of Graduate students in computer science for my advanced software engineering class. Since then I and a number of my colleagues have written several articles on this subject and have extended it into various domains: Variation-oriented engineering, variation-oriented requirements analysis, VOAD for SOA Solutions in addition to the base variation-oriented analysis and design. This topic also relates to my work on patterns for stability and symmetry in software architecture which I was writing for the pattern languages of programming conferences. I also did a blog entry on VOA that describes the high level basics of this paradigm.[More Current] VOAD is an integral part of the SOMA method for building end to end enterprise and applications solutions based on SOA principles (whether or not you have a full SOA implementation in mind. VOAD intersects with another very important notion that I have been researching and applying on projects, namely, context-aware services (CAS). Manners externalize semantics for on-demand composition of context-aware services.Manners define the context-aware behavior that is needed in order to externalize component manners to achieve greater maintainability and reuse. This article shows how you apply VOAD in order externalize the changing aspects of the software, so you can decrease maintenance costs, enhance reuse and actually increase flexibility in design and implementation of your software solution. This is a rather popular article and has been cited in a number of other publications . I think this is because we show the practical application of VOA to create a dynamically reconfigurable architectural style. Also, I believe this element is one of the characteristics that will be needed in more mature SOA Solutions. So as organization's adoption increases, they will need more mature ways of doing SOA, and VOAD is a key to achieving that.Empowering the business analyst for on demand computing explores Grammar-oriented object design (GOOD) which is the way in which we apply VOAD for externalizing manners. This article describes a project we did in the public sector for harmonization of multiple redundant business processes across the world. This is immediately reminiscent of the value that VOA brings: one of which is to consolidate business processes and optimize them by eliminating redundancy.I suspect that is why I am getting these emails on putting more out there on VOAD and to update my older work to reflect the impact and opportunities for SOA. I get the message :-) and will do so.
A new article in the IBM systems journal is out on SOMA that describes the method in a bit more detail than I did in 2004. This article describes a method that has now exercised hundreds of projects successfully and thousands trained worldwide in its delivery. This edition also features a number of articles on SOA.
Here is the abstract:
Service-oriented modeling and architecture (SOMA) has been used to conduct projects of varying scope in multiple industries worldwide for the past five years. We report on the usage and structure of the method used to effectively analyze, design, implement, and deploy service-oriented architecture (SOA) projects as part of a fractal model of software development. We also assert that the construct of a service and service modeling, although introduced by SOA, is a software engineering best practice for which an SOA method aids both SOA usage and adoption. In this paper we present the latest updates to this method and share some of the lessons learned. The SOMA method incorporates the key aspects of overall SOA solution design and delivery and is integrated with existing software development methods through a set of placeholders for key activity areas, forming what we call solution templates. We also present a fractal model of software development that can enable the SOMA method to evolve in an approach that goes beyond the iterative and incremental and instead leverages method components and patterns in a recursive, self-similar manner opportunistically at points of variability in the life cycle.
SOA is the next step in software engineering. The service-oriented paradigm comes to an increasing degree of maturity roughly 5 years after what I consider to be the start of serious corporate adoption.
We migrated from procedural computing to object-based and object-oriented computing. Then we moved to improve the state of objects with component-based software engineering. Now, we have evolved our best-practices as an industry and in academia and have moved into the era of service-oriented computing.
The gradual adoption and service maturity can be seen to move along a spectrum we have depicted in the Service Integration Maturity Model.Here we have seven levels of maturity across 7 dimensions from infrastructure to the business architecture.
The implication of this paradigm shift is that the best-practices of service orientation need to be integrated into the fabric of software engineering to be utilized on projects that (this is important, so you may want to mute the TV for a second) SOA may either not be part of it or a very small part of the project. Still, you would use SOA PRINCIPLES such as using the service model for a specification of the required functionality within a project or an enterprise.
What are your views on the use of SOA principles on non-SOA projects?[Read More]
In the world of SOA, the economics of funding determines what services actually get funded to be developed. This is why we need a set of criteria by which we can assess which candidate services are a priority for inclusion in your next budget cycle. We call these criteria or gating factors, the Service Litmus Tests (SLTs). SLTs are included in SOMA , the de facto end to end SOA Development Method .
Not all candidate services should be implemented as services; we may not have enough budget for that. But here this: we STILL need the functionality of the services that fail to pass the SLTs. Therefore those services still need to be implemented as part of some service component or realized by an application.
Economics is hand in glove with governance and of course SOA Gov includes the issues of economics, funding and budgets and ownership.
Ali_Arsanjani 120000D8QB 3,915 Views
Lots of SOA projects later, I have finally gotten back to this blog.
I am seeing a large surge in SOA projects, each with greater maturity and more complex needs; but some of the basic remain the same. Often, projectsface unpredictable and complex human situations which may defy rigorous algorithms and require the soft art of consulting. Some are attempting to bridle this erratic and unpredictable aspect of human and group behavior. More of the SOA Projects later.
A recent post by the Univ of Arizona talks about the development of a software that sifts through tons of data and "will use sophisticated computational methods based on game theory, co-evolution and genetic development models to find solutions that make sense in illogical times. Genetic algorithms analyze situations in an evolutionary context, where actions with the highest “fitness factor” (chance of achieving the greatest success) gravitate toward one another, produce offspring and eventually rise to the top."
This form of evolutionary, convergent behavioral computing promises to be used in more and more simulation situations.Typically, rule engines are cluncky (technical term) and finding relations between multiple rules is a tricky proposition.Patterns can help. For example, the Business Rule Pattern Language I have documented starts with a lighteweight way of handling rules and moves into increasingly complex ways of handling rules within object oriented applications.
Follwing its use on three past projects, I have recently revised this pattern language to support SOA. I have gotten quite a bit of email requesting this and I am responding by saying that I will be publishing a draft soon.
Ali_Arsanjani 120000D8QB 3,887 Views
As people stood in line sometimes a day before the big day, June 29, I was thinking of the partnership between Apple and AT&T. What about other Service providers who provide wireless services; could they not interoperate and have a device, which theoretically might be provider agnostic work with their phone was well? Of course economics, partnerships and politics are the key driving forces here...
So in the world of SOA, economics, partnerships and politics play a significant role as well. In the world of SOA, the economics of funding determines what services actually get funded to be developed. This is why we need a set of criteria by which we can assess which candidate services are a priority for inclusion in your next budget cycle. We call these criteria or gating factors, the Service Litmus Tests (SLTs). They are embedded in SOMA , the de facto end to end SOA Development Method.[Read More]
Ali_Arsanjani 120000D8QB 3,892 Views
SOA Solutions are most often hybrid solutions. Yes, they focus on a set of services; but they often do not soley rely on services for the realization of the functionality that needs to be in place for the business.
SOA solutions tend to rely on a combination of architectural styles and implementation and realization constructs to craft the underpinnings of an SOA solution. The SOMA method utilizes a combination of approaches to Service Identification. This includes 6 perspectives: top-down business process driven, business policy and rule-driven approaches, bottom up legacy integration, bottom-up legacy transformation (intrusive changes to rip out legacy modules and expose them via access points), information as a service typically used to consolidate multiple backend data stores and resolve inconsistencies in the access, rules and synchronization of the data stores and lastly the message-driven approach which seeks to integrate systems using a service interface.
Among the approaches above, although some are more established than others, information as a service affords a unique perspective in solving challenges relating to information. For example, data access interfaces and their underlying data access logic might need to be externalized judiciously if multiple channels are seeking to access and manipulate data from potentially multiple access points. The need for consolidation, synchronization and management of the data along with the need to have a coherent set of policies be applied to the data calls for the information service “entry point” to SOA.
Service-oriented modeling and architecture (SOMA) has become the industry de facto standard for SOA Methods. Introduced by IBM in Jan 2005, released recently as RUP/SOMA 2.4, it covers the identification, specification, realization and implementation of services, components, flows (processes), information and composition.
SOMA uses Information Analysis, Modeling and Planning during identification, an Information Specification during design and a number of artifacts during Realization and Implementation including considerations for Enterprise Information, Master Data Management, Conceptual, Logical and Physical Data Models.[Read More]