Welcome!

Microservices Expo Authors: Stackify Blog, Aruna Ravichandran, Dalibor Siroky, Kevin Jackson, PagerDuty Blog

Related Topics: Microservices Expo

Microservices Expo: Article

Considering the SOA Reference Model

Part 2 - Design Pillars

(SYS-CON Media) - The main drivers for SOA-based architectures are to facilitate the manageable growth of large-scale enterprise systems, to facilitate Internet-scale provisioning and the use of services, and to reduce the cost of organization-to-organization cooperation - SOA RM

When approaching a SOA implementation, I would like to consider two fundamental questions that many developers ask:
1)  What's the difference between service-oriented and service-based architectures?
2)  What special architecture elements are defined by the SOA RM?

In my opinion, the answer to the first is in the difference between the words oriented and based. I believe that smart IT organizations offer a lot of services already because the technical benefits of services have been well known for a while. However, the applications based on these services are still monolithic and don't provide convenient ways to implement business change. SOA, by contrast, elevates services to the level of first-class citizens by associating them (well, some of them) with business responsibilities. That is, business tasks can finally surface in the form of aggregated services that are open to business changes by design. SOA RM stresses a new role and meaning for services and outlines that:
1)  "The central focus of Service Oriented Architecture is the task or business function" versus both IT task/function and business operation automation task. This means that SOA primarily targets actual business services and processes instead of just improving existing (obsolete or legacy) applications.
2)  SOA service may be MUCH richer than a technical service because SOA service is supposed to run in the "execution context." The latter is defined as the set of technical and business elements that form a path between those with needs and those with capabilities. The "business elements" open absolutely new sphere of conditions, particularly business conditions that a regular technical service doesn't deal with.
3)  SOA "reflects business-motivating considerations" expressed "in service descriptions and service interfaces", where the service description can be expressed in the form of a contract. The service contract includes policies that "MAY also express business-oriented policies - such as hours of business, return policies, and so on".

If you elaborate on the business aspect of SOA, you'll get a much different picture of the service realm with different purposes, use, governance, maintenance, and, finally, architectural concept than we have in a "regular" service-based architecture. Besides these, I don't see any real differences. The service serves; the questions are how, where, why, when, to whom, and if it's needed after all.

The second question is trickier. The SOA RM standard doesn't articulate much on architecture elements. Instead, it defines SOA services and their relationship in new way that allows for addressing actual business services and processes by assigning business responsibilities to the services directly. That is, the standard defines what an IT service has to be oriented to in order to provide for IT agility to the business. I understand that the intention of SOA RM is giving us the definitions of the service and the "associates" that require a business service-oriented infrastructure. This infrastructure is the very architecture meant by the reference model.

So, since we're going to discuss SOA design pillars, let's list some commonly used elements of SOA infrastructure that include:

  • a SOA Service Communication Framework
  • a SOA Service Composition Framework
  • a SOA Service Transaction Framework
  • a SOA Service Security Framework
  • User Interfaces (including human UI) and
  • a SOA Service Component Adapter Framework
Elements of SOA infrastructure are reviewed in many technical publications; we will use them for references without deeper discussion.

Going forward, let's notice that some pillars relate to the SOA service design while others relate to the service interactions with other services and its consumers. Pillars 1 to 5 have ordered dependencies, the rest are applicable to the entire design process. Plus we'll use terms that have been defined and discussed in part one of this article, so you might look through it.

Service Design Pillars
Pillar 1
Considering Business Services selected for the implementation in SOA identify technical "vertical" and "horizontal" services. A vertical service implements one or a few business tasks while a horizontal service implements a supporting SOA infrastructure task, e.g. a Service Repository or Service contract management function. Traditional technical services like security, login, and the like also fit into the horizontal services category.

While the principle of separation of concerns is fully applicable here, the tasks implemented by the same vertical service have to provide "cohesive collections of behaviors" from the business perspective. Plus technical services have to be reasonably coarse-grain in their interfaces (to minimize the network round-trips and future modernizations in production caused by new technologies) but still provide flexibility. Nevertheless, one has to expect a certain amount of exploration and refactoring in this area of the service lifecycle.

When identifying horizontal services, first consider scalability, reliability, performance, expendability, and manageability - all together; consider programming convenience second. Services that are too generic usually don't provide the best performance just as too detailed ones cause problems in aggregation, management, and maintenance in light of frequent changes in the execution context.

Pillar 2
Considering business units of work, identify business data and its flows. Take into account two orthogonal sets of requirements - the presentation layer and the persistence layer requirements. Some units of work can require compositions of tem, so watch for a clear information transition with layered data format mappings. Also consider combinations of the business and just the technical data in the flows. That is, define metadata for the business and technical data upfront and operate at the metadata level (UML and WS-CDL can help).

Metadata is an important SOA element. Metadata provides the necessary data abstraction and allows for a data-quality controls. An XML Schema is one possible metadata definition and control instrument. This doesn't mean that only XML format is recommended. In some cases, especially for synchronous short-running transactions, programmatic Data Objects may be used for performance improvement. However, the Data Objects better be derived or generated from the XML Schema, which would provide metadata enforcement (related utilities are available in Java and C# and are known as XML binding).

Pillar 3
Define and lock down service interfaces and interfaces to the service resources such as persistence storage. Design the interface as business-oriented coarse-grained so your consumers can avoid frequent refactoring.

Best practices recommend avoiding RPC-like style in service interfaces. Web Services, RPC encoding, and SOAP-encoded arrays, in particular, seriously restrict service interoperability; consider a document/literal style instead. For non-XML cases, use a container style of interface to pass data structures and "exchanged" operations (aka commands in the Command Pattern).

It's important to design service interfaces with an eye to security, transaction, manageability, error handling, versioning, and interoperability from the beginning. Adding any of these things later is either extremely expensive (in effort, time, and money) or practically impossible.

For example, security is very important these days because of multiple technology threats and industrial/corporate security policies. Transaction support is dictated by its potential participation in service compositions. Manageability and error handling are core to any business service and process. Interoperability is one of the most critical factors in the SOA service lifecycle. That is, use of specific programming platform constructs can significantly reduce the utility of a service and lead to its refusal.

In spite of any ideas about keeping the service interface immutable, in real life variations in service invocation always happen over time. Anyway, a service interface isn't the only one to constitute a SOA service; there are also possibilities of SOA service behavior changes that affect service consumers. That is, the service has to be versioned. The versioning may be expressed via the interface and/or provided by a service registry. Only one condition has to be preserved in this regard - no service parts including the interface may be versioned separately from the service itself for service consumers.

Pillar 4
A service access protocol may be considered a part of the service interface definition or a separate issue. The spectrum of communication protocols used in SOA is wide including TCP/IP, HTTP/HTTPS, SMTP, and IIOP.

The designer has to choose communication protocol(s) for every SOA service. When considering a protocol, one has to take into account: 1) invocation model - synchronous or asynchronous; 2) the amount of transferred data; 3) performance; 4) extendibility; 5) scalability; 6) reliability; 7) transaction; and 8) manageability.

Protocols are also important for communicating with service resources. It's better to use standardized protocols where possible. For example, current best practice refers to the Web Services Invocation Framework (WSIF) as a binding bridge between Web Service's WSDL and a particular service resource protocol while Java components rely on JBI.

The right protocol is a balanced choice. For example, HTTP/SOAP-based protocol is slower than IIOP but it's more extendible; TCP/IP is more reliable than UDP but it's slower. Don't forget that one service may have more than one communication channel. For example, the same session EJB can be reached through RMI-over-IIOP and SOAP-over-HTTP/S.

The pioneers of SOA implementations have recommend doing a "protocol performance POC" for particular kinds of tasks. That is, the earlier in the design process a performance test gives the results the better for the entire project because refactoring in this area means costly redesign from the scratch.

Pillar 5
Define the SOA Service contract and Execution Context. One of the most important parts of the SOA service design is the definition of the SOA service contract template. While SOA RM simply says that "a contract...represents an agreement by two or more parties. Like policies, agreements are also about the conditions of using a service; they may constrain the expected real-world effects of using a service," the SOA service contract may be relatively complex. The contract, preferably written in XML, defines among other things:

  • runtime policies/constraints
  • runtime functional service parameters exposed to the service consumers; these parameters can be monitored and controlled
  • non-functional service parameters like scheduled service availability
  • business functional characteristics promised to service consumers like gathering audit information on service invocation events (the characteristics should be evidential and measurable)
  • all service interfaces and communication protocols exposed to a particular consumer
  • definitions or references to the Change Control procedures related to the service
  • service version information
  • references to the external (to your organization) SOA services invoked by the service. It's important for the business to know what other SOA services are involved because some political and business considerations may cause certain constraints on such involvement
  • some custom or business domain specific parameters if needed
The content of a SOA service contract isn't standardized. However it's the only document that's used by a cautious consumer to decide whether your service is the one that meets his needs. It is also the document that defines your liability as a service provider to your consumers.

Runtime policies better be derived from the SOA service contract. The policies have to follow standards wherever possible. As SOA RM defines it, "A service contract is a measurable assertion that governs the requirements and expectations of two or more parties. Unlike policy enforcement, which is usually the responsibility of the policy owner, contract enforcement may involve resolving disputes between the parties to the contract. The resolution of such disputes may involve appeals to higher authorities."

The SOA service implementation process treats the service contract as an objective set of development tasks. Not all implementation tasks ought to be expressed in the service contract, but only those that get visibility to the service consumers. Other tasks are the subject of regular application requirements.

Another fundamental characteristic of a SOA service is the execution context. According to SOA RM, the "Execution context is central to many aspects of a service interaction." The context includes concepts such as the SOA service contract, and behavioural and information models. The same service can be viewed and act differently in different execution contexts. For example, a logging service may store data "in the clear" when serving load-balancing calculations but it has to encrypt data for financial risk calculations (i.e., financial regulations constitute the execution context).

Pillar 6
Develop for security. SOA security isn't included in the SOA RM. Nonetheless it's obvious that the model's "real-world effects" can't survive without a business trust, i.e., without service security. We don't recommend developing a special implementation for each service. Instead use corporate security services at the level of service communication and service components, e.g., security policy syndication services.

Since a lot of security threats come from inside an organization, perimeter protection (external firewalls) isn't enough. We can outline the minimal SOA security requirements as 1) end-user bi-directional authentication with the service, and 2) inter-service bi-directional authentication. End-user and inter-service authorizations are much desired but may be optional in particular cases (e.g., a consumer may access everything after he's confirmed as a legitimate user). Other security measures are optional as well and depend on the nature of business data and processes.

Traditional security means like PKI, Kerberos, SSL, two-way SSL, user name and password tokens serve well in SOA. The service-specific means relate to the XML security standards. They include XML Digital Signature, XML standards for PKI (XKMS), SAML (for propagating the subject and its credentials), XACML (for transferring security policies and policy validation controls), and other encryption techniques applied to the exchange messages and communication channels. The choice of security methods to be applied depends on corporate and industry policies and regulations.

The takeaway here is that whatever security means ought to be implemented, they have to be considered in the design from the beginning. It's almost impossible or extremely expensive to add security in later phases of development.

Pillar 7
Develop to support transactions. SOA RM recognizes the importance of service transaction as "higher-order attributes of services' process models" but doesn't detail the topic. In practice, SOA uses commonly accepted standards for transaction management on different platforms and adds cross-platform transaction management standards. Transactions can be used as in an aggregated service (choreographed aggregation) as in an orchestration of services reflecting fragments of Business Process Management (BPM). The orchestration is typically based on the BPEL standard that introduces short-running and long-running transactions.

A short-running transaction is usually a synchronous RPC-type invocation running according to the known rules of transaction isolation. Transaction standards for Web Services and messaging include WS-Transaction and WS Reliable Messaging (WS-RM). A short-running transaction is good for performance and commonly used for co-located services or in cases where failure is unlikely and error handling is simple. A long-running transaction is an asynchronous transaction where the transactional state may be persisted for long periods of time (days, months, years) and are restored after a certain event. This kind of transaction is quite useful in manual operations or human decision making in the automated business process. BPEL also introduces a transaction capable of "undoing" the mistaken results of the previous transaction by compensating them (like an overcharge refund).

Transactional behaviour has to be embedded in the SOA service from the design phase. This will help provide future service aggregation- and orchestration-based SOA solutions that might not be clear at the moment the service is initially designed.

Pillar 8
Design for interoperability. As mentioned, interoperability is a crucial and critical characteristic of a SOA. The SOA RM says, "The value of SOA is that it provides a simple scalable paradigm for organizing large networks of systems that require interoperability to realize the value inherent in the individual components."

The existing WS-I Basic Profile address most interoperability issues in Web Services. CORBA IDL mapping does the same for the services communicating via IIOP. Communication via HTTP is strictly defined and doesn't leave room for interoperability issues (except for JavaScript flavors) while SOAP-over-HTTP/S still has some mismatch problems in mapping to different programming languages.

The recommendation here is to put a lot of attention on the data types and naming convention policies (besides the protocols) on the both sides of the interface - for the requester and the provider components. The policies must enforce compatibility and interoperability. One possible mechanism for data type and naming space interoperability control is XML Schema or any other schemas (metadata) accepted by the requester and the provider.

Another best practice recommendation is the creation of a proxy (isolation mapping layer) around the interfaces. That is, the data formats passed through the service interface shouldn't be driven by internal data formats (for either requester or provider). If the evolution of either side causes data format changes, it's better to create object mapping in the proxy than to try to modify the interface.

Pillar 9
Use Rules for flexibility and easy-to-change implementation. Rule engines are just tools used in SOA infrastructure implementation and certainly not a part of the SOA RM. However, being combined with policies on one side and decision-making points on the other side, rule engines become quite powerful instruments for providing a high level of flexibility and quick adaptation to business changes. Leading SOA infrastructure/framework vendors include different rule engines into their business process design tools and Enterprise Service Bus solutions.

While rule engines are really convenient instruments, there's no compatibility between engines from different vendors; a rule engine locks you into a vendor. Anyway, not everything should be driven by rules - rule maintenance is expensive and requires special skills. That is, the designer and developer have to be reasonable in using rule engines.

For example, if the business logic introduces a point for choosing from a finite number of cases (while cases change seldom), using rules may be overkill. Another known risk in using rules is a potential rule contradiction. That is, a rule engine generally doesn't automatically validate if a new rule contradicts any existing rules and any permitted combinations of those rules. Without such validation, the new rule can bypass restrictive rules and expose prohibited information (e.g., a junior clerk gets access to a high-volume account). So when using a rule engine in SOA, proceed with caution.


More Stories By Michael Poulin

Michael Poulin works as an enterprise-level solution architect in the financial industry in the UK. He is a Sun Certified Architect for Java Technology, certified TOGAF Practitioner, and Licensed ZapThink SOA Architect. Michael specializes in distributed computing, SOA, and application security.

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


@MicroservicesExpo Stories
How is DevOps going within your organization? If you need some help measuring just how well it is going, we have prepared a list of some key DevOps metrics to track. These metrics can help you understand how your team is doing over time. The word DevOps means different things to different people. Some say it a culture and every vendor in the industry claims that their tools help with DevOps. Depending on how you define DevOps, some of these metrics may matter more or less to you and your team.
For many of us laboring in the fields of digital transformation, 2017 was a year of high-intensity work and high-reward achievement. So we’re looking forward to a little breather over the end-of-year holiday season. But we’re going to have to get right back on the Continuous Delivery bullet train in 2018. Markets move too fast and customer expectations elevate too precipitously for businesses to rest on their laurels. Here’s a DevOps “to-do list” for 2018 that should be priorities for anyone w...
If testing environments are constantly unavailable and affected by outages, release timelines will be affected. You can use three metrics to measure stability events for specific environments and plan around events that will affect your critical path to release.
In a recent post, titled “10 Surprising Facts About Cloud Computing and What It Really Is”, Zac Johnson highlighted some interesting facts about cloud computing in the SMB marketplace: Cloud Computing is up to 40 times more cost-effective for an SMB, compared to running its own IT system. 94% of SMBs have experienced security benefits in the cloud that they didn’t have with their on-premises service
DevOps failure is a touchy subject with some, because DevOps is typically perceived as a way to avoid failure. As a result, when you fail in a DevOps practice, the situation can seem almost hopeless. However, just as a fail-fast business approach, or the “fail and adjust sooner” methodology of Agile often proves, DevOps failures are actually a step in the right direction. They’re the first step toward learning from failures and turning your DevOps practice into one that will lead you toward even...
DevOps is under attack because developers don’t want to mess with infrastructure. They will happily own their code into production, but want to use platforms instead of raw automation. That’s changing the landscape that we understand as DevOps with both architecture concepts (CloudNative) and process redefinition (SRE). Rob Hirschfeld’s recent work in Kubernetes operations has led to the conclusion that containers and related platforms have changed the way we should be thinking about DevOps and...
While walking around the office I happened upon a relatively new employee dragging emails from his inbox into folders. I asked why and was told, “I’m just answering emails and getting stuff off my desk.” An empty inbox may be emotionally satisfying to look at, but in practice, you should never do it. Here’s why. I recently wrote a piece arguing that from a mathematical perspective, Messy Desks Are Perfectly Optimized. While it validated the genius of my friends with messy desks, it also gener...
The goal of Microservices is to improve software delivery speed and increase system safety as scale increases. Microservices being modular these are faster to change and enables an evolutionary architecture where systems can change, as the business needs change. Microservices can scale elastically and by being service oriented can enable APIs natively. Microservices also reduce implementation and release cycle time and enables continuous delivery. This paper provides a logical overview of the Mi...
The next XaaS is CICDaaS. Why? Because CICD saves developers a huge amount of time. CD is an especially great option for projects that require multiple and frequent contributions to be integrated. But… securing CICD best practices is an emerging, essential, yet little understood practice for DevOps teams and their Cloud Service Providers. The only way to get CICD to work in a highly secure environment takes collaboration, patience and persistence. Building CICD in the cloud requires rigorous ar...
The enterprise data storage marketplace is poised to become a battlefield. No longer the quiet backwater of cloud computing services, the focus of this global transition is now going from compute to storage. An overview of recent storage market history is needed to understand why this transition is important. Before 2007 and the birth of the cloud computing market we are witnessing today, the on-premise model hosted in large local data centers dominated enterprise storage. Key marketplace play...
The cloud revolution in enterprises has very clearly crossed the phase of proof-of-concepts into a truly mainstream adoption. One of most popular enterprise-wide initiatives currently going on are “cloud migration” programs of some kind or another. Finding business value for these programs is not hard to fathom – they include hyperelasticity in infrastructure consumption, subscription based models, and agility derived from rapid speed of deployment of applications. These factors will continue to...
Some people are directors, managers, and administrators. Others are disrupters. Eddie Webb (@edwardawebb) is an IT Disrupter for Software Development Platforms at Liberty Mutual and was a presenter at the 2016 All Day DevOps conference. His talk, Organically DevOps: Building Quality and Security into the Software Supply Chain at Liberty Mutual, looked at Liberty Mutual's transformation to Continuous Integration, Continuous Delivery, and DevOps. For a large, heavily regulated industry, this task ...
Following a tradition dating back to 2002 at ZapThink and continuing at Intellyx since 2014, it’s time for Intellyx’s annual predictions for the coming year. If you’re a long-time fan, you know we have a twist to the typical annual prediction post: we actually critique our predictions from the previous year. To make things even more interesting, Charlie and I switch off, judging the other’s predictions. And now that he’s been with Intellyx for more than a year, this Cortex represents my first ...
"Grape Up leverages Cloud Native technologies and helps companies build software using microservices, and work the DevOps agile way. We've been doing digital innovation for the last 12 years," explained Daniel Heckman, of Grape Up in this SYS-CON.tv interview at 21st Cloud Expo, held Oct 31 – Nov 2, 2017, at the Santa Clara Convention Center in Santa Clara, CA.
The Toyota Production System, a world-renowned production system is based on the "complete elimination of all waste". The "Toyota Way", grounded on continuous improvement dates to the 1860s. The methodology is widely proven to be successful yet there are still industries within and tangential to manufacturing struggling to adopt its core principles: Jidoka: a process should stop when an issue is identified prevents releasing defective products
We seem to run this cycle with every new technology that comes along. A good idea with practical applications is born, then both marketers and over-excited users start to declare it is the solution for all or our problems. Compliments of Gartner, we know it generally as “The Hype Cycle”, but each iteration is a little different. 2018’s flavor will be serverless computing, and by 2018, I mean starting now, but going most of next year, you’ll be sick of it. We are already seeing people write such...
Defining the term ‘monitoring’ is a difficult task considering the performance space has evolved significantly over the years. Lately, there has been a shift in the monitoring world, sparking a healthy debate regarding the definition and purpose of monitoring, through which a new term has emerged: observability. Some of that debate can be found in blogs by Charity Majors and Cindy Sridharan.
It’s “time to move on from DevOps and continuous delivery.” This was the provocative title of a recent article in ZDNet, in which Kelsey Hightower, staff developer advocate at Google Cloud Platform, suggested that “software shops should have put these concepts into action years ago.” Reading articles like this or listening to talks at most DevOps conferences might make you think that we’re entering a post-DevOps world. But vast numbers of organizations still struggle to start and drive transfo...
Let's do a visualization exercise. Imagine it's December 31, 2018, and you're ringing in the New Year with your friends and family. You think back on everything that you accomplished in the last year: your company's revenue is through the roof thanks to the success of your product, and you were promoted to Lead Developer. 2019 is poised to be an even bigger year for your company because you have the tools and insight to scale as quickly as demand requires. You're a happy human, and it's not just...
"Opsani helps the enterprise adopt containers, help them move their infrastructure into this modern world of DevOps, accelerate the delivery of new features into production, and really get them going on the container path," explained Ross Schibler, CEO of Opsani, and Peter Nickolov, CTO of Opsani, in this SYS-CON.tv interview at DevOps Summit at 21st Cloud Expo, held Oct 31 – Nov 2, 2017, at the Santa Clara Convention Center in Santa Clara, CA.