McKinsey: Why digital transformation should be a priority for health insurers

McKinsey has a great article on the impact of the digital revolution on the health insurance sector:

http://www.mckinsey.com/industries/healthcare-systems-and-services/our-insights/why-digital-transformation-should-be-a-strategic-priority-for-health-insurers (May, 2016)

Market Changes in Health Insurance

Digital innovation in health insurance can

• achieve substantial SG&A cost savings (10-15%)
• decrease medical spending (lower-cost alternative services; customers are the major beneficiaries)
• drive organizational transformation
• change the operating model
• improve customer engagement and satisfaction

Disruption starts to impact health insurance

• healthcare has traditionally been slow on customer centricity
• digitally native firms are entering the market and could severely disrupt
• incumbents must react now to avoid irrelevancy
• VC investment in digital health is 4.5B p.a. (up 400% in 4 years)

Compared to other sectors

• heavy regulation is no protection for incumbents (compare financial sector)
• disruption takes about 5 years from tech innovation to industry change (compare music industry)
• P&C: disruptors lead with a direct-distribution model instead of broker-led sales

Changing customer behavior

• Consumers have become accustomed to online and mobile channels and will expect these from health insurance as well (during purchase and service)
• Incumbents shift IT budgets from core to digital capabilities (50% within 3-5 years)
• Tipping point for market shift when a) consumer change becomes widespread and b) network  effects take hold 

Digital Impact relies on Four Levers

Four levers (and their combination) can yield significant improvements to health insurance:

Stronger connectivity: better engagement with all stakeholders:

• For customers: more frequent interaction on the basis of personalized, omnichannel services that optimize care and allow for better health management.
• For providers: improved collaboration & coordination, optimized care delivery, better health population management.

Greater efficiency and automation:

• radically reimagine workflows to align with the health journeys of customers and other stakeholders
• e.g., a clean-sheet customer-centric online sign-up process with reduced escape rate and cycle time (as done by one health insurer)

Better decision making:

• Analytics helps to better understand customers & risk
• Optimize population health management
• Innovate around payments & products (e.g., customer segment focus)

Business Model and Care Innovation (Examples):

• Wearables monitor patients with chronic conditions
• Telemedicine, “virtual visits”
• Locate nearby providers
• Easy access to medical history 

Recommendations

Customer Journey Redesign

• Identify customer journeys around enrollment, billing, claims, submissions/processing and issue resolution; in detail:
• Increase consumer awareness: product education and evaluation
• Enroll customers: inquiry, quote-to-card, billing
• Solve Problems: for customers, brokers and providers
• Pay what’s owed: claims processing, broker commissions
• Analyze and report data: broker reporting, etc.
• Help members take control: care management
• Help providers deliver better care: performance review
• Optimize the provider network: network design, credentialing
• Prioritize and analyze customer journeys; risk and stakeholder assessment
• Redesign and digitize to the fullest extent possible where value is high and risk is low

Organizational Redesign

• Break functional silos by creating cross-functional teams (XFT)
• include multiple business segments, IT and operations (BizDevOps)
• XFT teams work collaboratively on customer journey redesign and bring in different perspectives
• Leaders should set bold quantitative targets (e.g., 50% improvement in cycle time)

Resource Allocation

• Shift budgets from run to change to sustain the digital transformation
• Focus on a small number of high-priority journeys
• Focus on talent in the digital transformation team

Bimodal IT and Decoupling

• Adopt a bimodal IT approach to enable rapid digital innovation
• Integration can decouple the innovation layer (which supports minimum viable products, fail-fast, quick development times) from the stable core
• Integration should standardize around data models and functional services
• Standardize existing data models within the legacy infrastructure
• Create reusable standard service interfaces to avoid point-to-point integration

Six step plan for Incumbents

1. Start from the customer: Customer Journey end-to-end digitization
2. Break functional silos: cross functional BizDevOps teams with clear mandate
3. Create measurable targets for each team
4. Resource allocation and budgets: reallocate investments
5. Focus on talent: get and retain digital talent
6. Maximize value of two-speed IT: digitally enable legacy infrastructure

The ITMC Take

McKinsey describes the business case of digitization for the health insurance sector in terms of bottom line results and improved customer satisfaction. The bigger picture however is a win-win situation for health insurers and their members: Business model innovations have the potential to improve the quality of care while driving down medical cost. While the details are not clear yet, the impact of innovations can be enormous. Analytics, wearables, connectivity work in concert to steer members towards low-cost prevention and a healthy lifestyle, create awareness and enable them to better manage their health through personalized recommendations.  Some customers and many providers may cringe at this prospect. But the rapid consumerization in healthcare through wearables demonstrate that many customers want to take responsibility for their health. Health insurance must develop the necessary digital capabilities to respond to future market developments now. 

WebLogic JMS in-flight Message Compression

WebLogic JMS has a feature for compression of in-flight JMS messages. Compression happens between the JMS client library and the JMS server. This avoids bandwidth issues and nasty problems with maximum message size limitations (see http://thomasrischbeck.blogspot.ch/2012/03/jms-maxmessage-size-persistent-messages.html).

Dynamic compression can be enabled on the JMS connection factory. Here you specify a threshold in bytes. If the serialized message body exceeds this threshold and if the client is not collocated then the message payload is compressed in-flight. See Compressing Messages in the WebLogic Performance and Tuning Guide.

An alternative is to programmatically set a compression threshold via WLMessageProducer interface.

Interactive OSB Testing

Interactive Testing with OSB Test Console

Oracle Service Bus has a built-in test console for the interactive testing of OSB artifacts at runtime. It generates synthetic payloads (including security tokens) based on the service description. Note that the test console performs an actual invocation of services - which can have unintended consequences on production systems. Therefore, use the test console only on development and staging systems and block it on production systems. 

Note that interactive testing only gets you so far. What you really want is automatic unit testing of OSB as the basis for regression testing and continuous integation, as well as test-driven development (TDD).

The test console is accessed via the icon 

Business Service Testing

Business services are backend service definitions in OSB. The test console invokes the actual backend service and therefore tests its correct endpoint configuration, its availability as well as any downstream dependencies the backend service may have (DBs, etc.).

Proxy Service Testing

Test console can invoke proxy services. The invocation triggers off the proxies pipeline which is graphically displayed in an invocation trace . The invocation trace contains detailed information about each step and variable values recorded during execution.

Testing Transformations

The following transformation types are supported in OSB an can be tested with the test console. For various transformation actions the test icon is available:

• XSLT transformations in OSB are used for XML-to-XML mappings. You can test them with source XML formats within the test console. The test console displays named parameters supported by the transformation.
• XQuery transformations  work with either XML or non-XML input
• MFL is used to describe non-XML data formats and the layout of binary and text data in these formats.  The test console can verify the correct bidirectional mapping.

JMS MaxMessage Size - persistent messages

Large JMS Messages in WebLogic

 

Applies to	WLS 10.3.5 / OSB 11g - plain vanilla setup, 64bit, 3GB Heap
Requirement	Send large messages (250MB) across OSB using JMS Transport
Background	OSB supports streaming mode for HTTP and File-based protocols - which we tested successfully with multi-GB files. However, things are not so straightforward with JMS proxies which essentially boils down to the WLS JMS subsystem.
Test Setup	The test is based on a JMS client that creates a persistent message with a variable payload size (1-250 MB). It sends this message to a JMS queue on a remote WLS server and receives it synchronously from the same queue. We measure the roundtrip time on the client as well as server load (heap, CPU) on the WLS server.

 

First Run

 

 

At first it looks good - you'd get a near linear increase in response time in relation to the  message size. But then - at 10MB size - you will encounter the first roadblock:

 

weblogic.socket.MaxMessageSizeExceededException: Incoming message of size: ’10000080′ bytes exceeds the configured maximum of: ’10000000′ bytes for protocol : ‘t3′.

 

Increase MaxMessageSize

 

For most protocols, including T3, WLS limits the size of a network call to 10MB by default (see Performance and Tuning Guide). The 10MB limit is a soft limit meant as a protection against DoS-attacks. It is worth noting, that a server is more prone to DoS attacks when increasing this limit. Anyway, let's increase it to see how large a message we can process.

 

• Go to Managed server--> Protocols --> General and set the Maximum Message Size to something large.

• To set the maximum message size on a client, use the following command line property: -Dweblogic.MaxMessageSize

• Verify the Queues maximum accepted message size:

Queue-> Configuration-> Threshholds and Quotas-> Maximum Message Size

 

 

 

Setting Maximum Message Size for Network Protocols

 

 

Increase Server Heap Size

When running the test again you'll notice that the server-side heap is really busy on those large messages. 

 

 

 

Out heap size of max 3GB seems to be a limiting factor. So we increase it to 8GB (note: you need a 64-bit JVM for anything larger than 3GB):

 

USER_MEM_ARGS="-Xms3096m -Xmx8192m -XX:PermSize=128m -XX:MaxPermSize=512m -XX:NewRatio=3 -XX:SurvivorRate=128"

 

And off we go with our next test.  This time we can send up to around 100MB messages. Roundrip times increase up to 50-80 seconds.  Then we get a CORBA exception:

 

weblogic.jms.common.JMSException: weblogic.messaging.dispatcher.DispatcherException: java.rmi.MarshalException: CORBA COMM_FAILURE 1398079691 Maybe; nested exception is:

org.omg.CORBA.COMM_FAILURE:   vmcid: SUN  minor code: 203 completed: Maybe ...

 

 

Bypass CORBA Stack

At this point we got a hint from Oracle support: "Switch from wlclient.jar to wlthint3client.jar so you bypass all the CORBA stack…". This change guarantees that the T3 protocol is used for JMS communications.

 

We can now send messages up to around 220 MB of size - albeit at extremely long roundtrip times of around 10 minutes. Try anything larger and you get:

 

weblogic.jms.common.JMSException: weblogic.messaging.dispatcher.DispatcherException: weblogic.rjvm.PeerGoneException: No message was received for: '240' seconds

at weblogic.jms.dispatcher.DispatcherAdapter.convertToJMSExceptionAndThrow(DispatcherAdapter.java:116)

at weblogic.jms.dispatcher.DispatcherAdapter.dispatchSyncNoTran(DispatcherAdapter.java:61)

at weblogic.jms.client.JMSSession.receiveMessage(JMSSession.java:894)

at weblogic.jms.client.JMSConsumer.receiveInternal(JMSConsumer.java:647)

at weblogic.jms.client.JMSConsumer.receive(JMSConsumer.java:511)

 

 

 

In-flight Compression

 

WLS JMS has another nice feature: online compression of messages during transport (see

Compressing Messages). Since our test messages contain EDIFACT sample payloads they can be reduced to around 10% of their original size.  This also has a dramatic effect on roundtrip times, as  the chart below shows (up to 100MB). 250 MB messages can now be easily transferred and take less than 10 seconds for roundtrips.

 

 

 

Conclusion

We didn't crack the 200MB limit on physical message sizes in WebLogic JMS. However, we found a solution to our particular requirement. It would be interesting to find out, how to get rid of the PeerGoneException in Test-03. Even though, the roundtrip times for this kind of message size are prohibitively long.

Oracle ADF Powerkurs - 15. Dezember in Bern

Am 15. Dezember findet in Bern ein 1-tägiger Powerkurs zum Thema Oracle ADF Entwicklung statt. Durchgeführt von Frank Nimpius, dem Experten der Oracle für das Thema. Weitere Details und Anmeldung unter: http://www.ipt.ch/veranstaltungen/javaee-applikationsentwicklung/.

JavaEE-Applikationsentwicklung

Schnell und effizient mit Oracle ADF

Am 15. Dezember 2011 in Bern

Suchen Sie ein effizientes Framework für die Entwicklung Ihrer Rich-Enterprise-Applications im Web- und Mobile-Bereich? Oder planen Sie eine Rapid-Application-Development-Initiative in Ihrer Organisation? Wir zeigen Ihnen an diesem Event, wie Sie mit Oracle ADF signifikante Effizienz- und Qualitätssteigerungen in der Java EE Applikationsentwicklung erreichen können. Ergreifen Sie die einmalige Gelegenheit Frank Nimphius persönlich kennenzulernen, den Experten für Oracle ADF im deutschsprachigen Raum.

 

Nach einer Einführung in die Herausforderungen bei der Applikationsentwicklung zeigen wir Ihnen, wie das auf Standards basierende End-to-End Framework Oracle ADF aufgebaut ist. Dann vergleichen wir es mit anderen Frameworks für die Applikationsentwicklung und zeigen auf, wie Oracle ADF Sie bei der Realisierung der Rapid-Application-Development-Konzepte unterstützt.

Die Demo nach der Mittagspause setzt die am Morgen präsentierten Ansätze praktisch um und zeigt wie schnell eine Rich-Enterprise-Application mit den visuellen und deklarativen Werkzeugen der ADF-Plattform entwickelt werden kann.

First-class citizens and splendid isolation

Michael Poulin did a review of our SOA Design Patterns book:

Thomas Rischbeck says: “The ESB pattern is about performing integration tasks and adding value to client-service communication in an SOA – all completely transparent to the participants”. I would welcome this transparency if it were not based on intelligent routing and data enrichment having its own policies on the interaction between the participant. In SOA, we have a mechanism that regulates the interactions – Service Contract – and it is not seemed to be considered by the ESB at all.

Michaels seems to imply that the ESB doesn't honor service contracts at all. I get the impression that Michael also sees an ESB as a magical integration switchbox in the middle of the enterprise. This makes Michael feel uneasy – which I can understand very well.

There’s also another way of looking at an ESB: You could think about it as an opaque substrate for hosting "facade" services – which are fully-fledged, first-class citizens in the SOA world. Like every other service, such facade services have a formal interface (and a less formal contract which is a superset of the interface (as James Watson says. While I’d admit that there can be quibbles about business logic on the ESB – it doesn’t “break […] the service contract between the service consumer and the service provider”, as Michael thinks. The consumer wouldn’t’ reason about the actual backend provider but only about the ESB-exposed endpoint it communicates with. The rest is implementation details, so to say.


From this point of view the ESB becomes opaque. It cannot violate a service contract because it isolates whatever backend services are multiplexed through the ESB. This splendid isolation renders numerous benefits. For example, services can be normalized according to enterprise requirements; they may also enforce coarse-grain security and shield backend services from certain attack vectors. Many rings of defense is a security best practice since the middle ages.

Cheers, Thomas.

Ride the Cloud at Lightning Speed

How can you make your computationally intensive application run like lightning on the cloud? Use Cloud Parallel Processing!

Some computational tasks are so demanding that a single machine cannot process them within a reasonable amount of time. Applications that leverage massive amounts of data from the Web, a social media network or any other large-scale system can easily overwhelm a single machine. Many other examples exist of analytical tasks that operate on large data sets.  

Parallel processing has been automated and widely practiced at the processor level. In order to utilize the cloud efficiently it is necessary to scale out. Scaling out means that multiple nodes share the workload and therefore get things done faster. Parallel processing at the application level is necessary to harness the power of the cloud.

Creating a Scalable and Elastic Application

With the advent of cloud computing a large-scale and virtually unbounded grid of compute nodes becomes a reality. In principle, this makes the cloud an ideal substrate for on-demand high-performance parallel processing. The cloud can adapt elastically when CPU requirements change and spin-up or tear-down nodes accordingly.

In the latest Gartner Hype Cycle for Cloud Computing (July 2010), Cloud Parallel Processing is an entrant of the Cloud technology adoption curve. They note that “as cloud-computing concepts become a reality, the need for some systems to operate in a highly dynamic grid environment will require these techniques to be incorporated into some mainstream programs. [] The application developer can no longer ignore this as a design point for applications.”

Parallel Programming is hard!

This points to the caveat of parallel processing: it is very hard to build scalability and elasticity into general purpose applications. But if you cannot make multiple nodes share the workload then the cloud will not help you get things done faster.

How can you make your computationally intensive application run in parallel on several nodes? A parallel programmer must address these steps:

•          Partition the workload into smaller items
•          Map these onto processors or machines
•          Communicate between these subtasks
•          Synchronize between subtasks as required

Note that synchronization is a real bugger – because it is anathema to parallelism. Yet it is unavoidable in most applications. 

Shift of Programming Model

Even with the use of toolkits that offer communication and synchronization primitives (such as PMI) parallel programming is a bit like rocket science. Without a shift in programming model parallel programming will stay elusive. Programmers will have to deal with parallelism to some degree at least – they can no longer rely on middleware, databases and operating systems to exploit parallelism.

Chris Haddad mentions the following programming model shifts that happen with the cloud:

•  Actor Model: Dispatching and scheduling instead of direct invocation Queues and asynchronous interactions
• RESTful interactions Message passing instead of function calls or shared memory
• Eventual consistency instead of ACID

Fortunately, parallel programming has been a research topic for some decades already – long before the cloud existed. Demand for high performance computation stemmed from various domains, such as weather prediction, finite element simulation and other data analysis. During my PhD I worked on object-oriented parallel programming using COTS (commercial-off-the-shelf) workstation clusters. These are called distributed memory machines in parallelism jargon. The cloud is nothing else than a distributed memory machine - albeit at a much larger scale. And my work also touched the Actor model that Chris mentioned. 

Watch this space

The principles and models that apply to distributed memory machines would also apply to the cloud (essentially, the cloud is a very large distributed memory machine). In a follow-up series of posts I will look at those topics in more detail and examine how they can be applied to Cloud Parallel Processing. I’ll dig deeper into programming models shifts (such as Actors) and outline some parallel algorithms (such as map reduce, n-body and others of the 13 computational dwarfs). 

Welcome to Giovanni Scheepers - Asset Liability Management for SOA

Welcome to the Blogosphere to Giovanni Scheepers! Am looking forward to a good discussion and espeically looking at your idea of asset liability management in the context of SOA. Am very excited!

Offroad Google

While mountain-biking in the Swiss alps we met this man recently. What looked like an ice cream bike from the back turns out to be an e-bike with Google equipment mounted. I'm really curious to see these pix show up on Google street view (or trail view).

SOA Symposium 2010 in Berlin - CFP

The SOA Symposium 2010 is taking place in Berlin this year. The Call for Papers is open until June 30. Themes and topics this year:

Exploring
Modern Service Technologies & Practices

•   Service Architecture & Service Engineering

•   Service Governance & Scalability

•   SOA Case Studies & Strategic Planning

•   REST Service Design & RESTful SOA

•   Service Security & Policies

•   Semantic Services & Patterns

•   Service Modeling & BPM

Scaling
Your Business & Infrastructure Into the Cloud

•   The Latest Cloud Technology Innovation

•   Building and Working with Cloud-Based Services

•   Cloud Computing Business Strategies

•   Case Studies & Business Models

•   Understanding SOA & Cloud Computing

•   Cloud-based Infrastructure & Products

SOA Forum 2010 - Asset-based SOA

A total of 203 attendees met at the annual SOA Forum in Zürich last week. The pre-conference day was made up of four half-day sessions around the topics of SOA Security, BPM, Business Service Modelling and SOA Testing. 175 attendees came for the conference main day. The topic of this year's event was the industrialization of SOA based on reusabel assets.

As keynote speaker Massimo Pezzini of Garnter pointed out: assets come in many flavours. They can be infrastructure (SOA backplane), documentation, process templates, organizational templates, skills ... and people! Massimo differentiated three different SOA-flavours: RPC, WOA (request-driven, REST) and Event-driven SOA.

Keynote speaker Anne Thomas Manes talked about the reincarnation of SOA. 18 months ago Anne published a eulogy on SOA. This eulogy wasn't praise more than lamentation at the demise of a promising development - that met it's untimely dead because of hype and wrong application. A talent wasted?

Well - SOA's resurrection happened. Although the connotation here is somewhat religious. Anne suggests to avoid all religious and fundamentalist mindsets when applying technology and architecture. SOA is a logical choice and sine-qua-non for many problems. But it's not a panacea. An opinion that was reflected in the panel discussion.

The wrong approach to SOA - according to Anne - consisted of too strong a technology focus (ESBs, appliances, WS-stacks, governance tools and so on) and too little focus on architecture. Frequently systems were built with an integration mindset which is akin to putting lipstick on a pig (Anne has a lovely picture in her slidedeck). Ultimately this led to high maintenance costs. According to her figures the development of a system is only 7-15% of the TCO - the rest is maintenance. IT budget cuts mean that innovation comes to a screeching halt because there is not much opportunity to reduce maintenance costs.

The reincarnation of SOA consists of seeing the big picture rather than technologies and point solutions. She recommends to learn from the Web 2.0 technologies and help enterprise software break out from the firewall. Externalization, consumerization and democratization of IT are the keywords here. A sentiment echoed by Zapthink and others.

SOA's reincarnation is driven by the application of design principles and architecture instead of focussing on technology. SOA principles are applied through:

• Business Services Modeling
• SOA Design Principles

When these principles are applied correctly a system should exhibit the following characteristics:

• Modular
• Distributable
• Discoverable
• Swappable
• Shareable

The SOA Forum presentations can be accessed under the following URI:

www.ipt.ch/downloads

 Username:    soaforum_main

Passwort:      assets

ESB Book - New Cover

The book "Modern ESB Architecture for SOA" I am writing together with Mark and Arnaud now has a new and final cover.

SOA Forum 2009 is history

On Wednesday the SOA Forum 2009 (http://www.soa-forum.ch) came to an end. We had a spectacular crowd of 175 attendants. Below are the slides that I presented in Track 2 together with Martin Kräuchi:

End-of-time Prophecies on Middleware, SOA, ESB and else

When Nostradamus pronounced and published his famous prophecies in the 16th century he did not need to worry about target consumer categories. His predictions - some of them interpreted as end-of-time prophecies or just plagues, earthquakes, etc. - where all-encompassing. Unfortunately, none of his writing was ever interpreted as predicting an event before it actually ocurred. :-)

So, what about those target consumer categories? I came across these when reading an excellent article by Steve Vinoski with the title "Is Middleware Dead? (IEEE Internet Computing, vol. 8, no. 4, 2004, pp. 94-96). This article is also about predictions -- in the technology and market domain this time.

Steve Vinoski wrote this article in response to "The End of Middleware", where Jonathan Schwartz of Sun touted middleware to be dead. The assumption behind this pronounciation: everybody would use Sun's Java Enterprise System then there would be no need to integrate with other middlewares. Of course, it did not pan out this way.

Similar pronouncements about the death of CORBA, SOA, ESB and many other technologies have been made. Steve points out that such predictions frequently do not take the target consumer category into account. According to Crossing the Chasm, there are five such categories in a technology adoption lifecycle (- innovators, - early adopters, - early majority, - late majority, - laggards).

When an "early adopter" feels that a certain technology is dead -- it may be alive and kicking from the point of view of the "laggard". One-size-fits-all solutions are therefore impossible. Where are ESBs on that adoption bell curve? Pretty much at (or slightly after) the peak, I'd say.

Stateful vs. Stateless Services

From a runtime point of view it is advantageous to have stateless services. Statelessness - from a conceptual point of view -- means that the service does not "remember" state between consecutive invocations. Of course, the service may still store information in the DB backend. The important point, however, is that this state is not within the service instance or the service volatile memory.

Clearly, stateless services do scale better: Additional instances of the service can be fired up to cope with increased load. They can also deal better with failover situations. A client request may be routed to an alternative instance transparently when the original instance goes down.

However, sometimes stateful services do make sense from a modelling perspective. Nicolai Josuttis, for example, describes a shopping cart service. For this type of service, a client would expect to successively call operations to add new items.

A stateful service may be implemented as a stateless service: either by sending all relevant information (eg the shopping cart contents) with every request or by sending a session ID with every request and keeping the shopping cart contents in the DB. Or a "real" stateful service keeps the shopping cart state in memory - this requires session affinity: ie, each successive request by a given client must be routed to the same service instance. When this instance goes down, the user must start their shopping trip all over again -- possibly an acceptable tradeoff in an eCommerce application. Nicolai Josuttis works out the associated tradeoffs (state in frontend, state in service, state in backend) really nicely in his book (p.195).

Oracle aims for a Slice of Cake

Frank Martin of Oracle quipped that after the takeover of Sun, Planets and Moon might follow soon. Meanwhile, The Economist had another very good high-level story about the Sun takeover by Oracle titled "Mr Ellison helps himself". Great analysis with the conclusion that our industry is going from horizonal/modular to vertical/integrated with the Sun acquisition being just one indicator of the overall trend. The cake (market) is more segmented into slices than into layers. This has significant repercussions regarding the IT value chain that all players wil need to adapt to.

Message-Driven SOA = Robustness

I love Bill Poole's blog - In a post last year, Bill describes a service model that is built around cohesive business areas/capabilities and along the following principles: large grain services; asynchronous reliable communication; no centralized data; no cross-service transactions.

In the past I've worked as architect on an eGovernment project that followed just these same architectural principles. That was by design and also by necessity -- since the underlying product (SonicESB) naturally guided developers into modelling services as entities with a purely message-oriented interface (of course, that was putting the horse before the cart). Services in SonicESB aren't really services in the conventional sense -- they are written against an API and can communicate with other Sonic services only using that API. The API binds onto a JMS abstraction layer that exercises the underlying MOM implementation. However, some queues that drive services may be exposed via Web service interfaces to the external world. Sonic also offers a light-weight process construct: itinerary-based routing includes a routing slip with every message and describes which queues/services must be traversed in what order.

The project team really bought into the benefits of temporal decoupling and came to see it as godsend -- they also came to accept the unusual asynchronous communication style that goes with it.

• individual service may be taken down for maintenance without negatively affecting the rest of the system -- messages simply get buffered until the service comes up again
• as a corollary to the above: services do not need to handle peak loads -- just the average load. When bursts of messages arrive they are simply backlogged in the queue until the service can process them.
• non-blocking for the message producer: fire-and-forget with reliable messages; producers "know" that messages eventually arrive - SLA of the recipient is can help in giving a deterministic timeframe
• monitoring of queues: analyze queue utilization over time; detect usage patterns and adapt to and/or predict load situations;
• inspection of in-flight messages: analysis of messages; remove and store individual messages; replay them at a later stage (or for testing); isolate poison messages and use them for failure analysis

While this particular system was built on top of Sonic, the general principles can be carried over onto any technology that supports reliable asynchronous messaging. Examples are JMS, AMQP, XMPP, WS-Notification, and others. The problem here is: how do you schema-describe your messages? This can be done by out-of-bounds mechanisms, albeit unsatisfactorily.

Virtualization at the Endpoint?

Aaron Skonnard describes the Microsoft Managed Services Engine (MSE) over at MSDN.

Service virtualization is nothing new, of course - it's a pattern that underlies most SOA middleware products (ESBs) and agents (Actional, AmberPoint,..). Hosting platforms (WCF, App Servers) also provide rudimentary virtualization on top of the business logic.

The strength of virtualization lies in its power to isolate consumer and provider and decouple them -- more so than what would be possible if they interacted directly (somebody once said - not without irony - that "any problem can be solved by another layer of indirection" - was it an R. Johnson?). This shields a service developer from the requirements of the consumer and allows her to focus on business logic -- rather than getting multiple stakeholders to agree on the non-functional aspects of the service interface.

ESBs place service virtualization "in the middle" - the MSE that Aaron Skonnard describes places it at the endpoint. Endpoint-based virtualization is becoming a new trend. WCF does it and some other containers - to a limited degree. Some LOB packaged applications start to offer this functionality at the endpoint (e.g., Siebel with the inclusion of Fusion Middleware). This is also in support of pure SOAP architectures that describe the enterprise as just a collection of SOAP-speaking endpoints (with reliability, security, etc. all being controlled through WS* protocols).

Virtualization at the endpoint, however, is not equivalent to virtualization in the middle. It limits you to those aspects of a communication that occur -- well, at the endpoint -- just before hitting the business logic. All aspects that are more centralized conceptually, are more difficult to address. Think about location transparency, for example. If this aspect is implemented on a central, shared intermediation layer it is simple to move backend service providers to new locations. If there is no such infrastructure, then every consumer must be able to directly address the provider endpoint. Either you configure these addresses in peer-to-peer fashion (which is a mess) or you do some kind of lookup: this could be via DNS (only works with HTTP), a registry or a centralized resolver service. The two latter solutions require that the consumer endpoint (either the stack or the business logic) handles the address resolution.

Then, think about content-based routing -- and things get more difficult. This would be a functionality that the consumer endpoint would provide -- or an intermediate infrastructure service. Other examples for virtualization/intermediation requirements taht map more easily onto a (conceptuall) centralized implementation, rather than onto a set of distributed smart endpoints:

• messaging patterns (pub/sub, reliable, async) - reliable queueing can be done at the endpoint but it's a maintenance nightmare (think trapped messages, poison messages, ...)
  
• complex event processing - for correlation/pattern detection, all events must be brought together at the event correlation engine for the application of rules
  
• message routing (addressing, CBR) - as mentioned above
  
• layered security defense - combination of multiple security mechanisms where threats are filtered out at multiple policy enforcement points (PEP) along the communication path
  
• control point for runtime governance (can you really rely on your distributed containers?) - to some degree, since you can monitor messages in progress, but not the availability of services that are not in use
• Provididing multiple interfaces to different clients at different network locations - maybe you only want to make a certain "engineered" interface (different operations, different policies) available to another department, DMZ-external, etc.

The following diagram gives a conceptual idea of where intermediation/virtualization can take place when looking at the consumer-provider communication path:

Developer Summit 2009 - Stockholm

I'll be speaking together with Herbjörn Wilhelmsen at the Developer Summit 2009 in Stockholm about How to Survive the Bermuda Triangle of Legacy, Services and ESB.
Developer Summit, Scandinavia’s leading developer conference is turning 5 years! This year 40 sessions on 6 tracks, key-notes and workshops are planned.

SOA Forum Schweiz 2009

The SOA Forum 2009 will take place May 12-13 in Zurich Sihlcity's cinema mediaplex. The motto this year: "Beyond the hype and theoretical discussions. Use SOA right – succeed in practice”.

Be sure to mark the date in your calendars! We'll have some really interesting and controversial keynote speakers at the start this year - wrapped up with a panel discussion.

• Thomas Erl: "Next Generation SOA"
• Anne Thomas Manes (Burton): “What Becomes of SOA During Though Economic Times?”
• Michael von Uechtritz (Gartner): final keynote

I'll do a talk on mediation strategies in SOA and have a look at a SOA reference architecture.