Listen to the entire 8:53 podcast Download file

If the idea of carving up an SOA strategy sounds a little ghastly, ebizQ is scaring up some help this Halloween with the SOA in Action Virtual Conference. One of the keynotes for the conference, scheduled for October 31, is "Building a Strategic and Tactical Platform for SOA" with Forrester Research's Randy Heffner. Heffner joined ebizQ's Gian Trotta recently to offer a preview and some quick tips on SOA strategy.

SOA Misconceptions (or Tricks)

Web services and SOA are not the same thing, says Heffner, and viewing the two concepts that way tends to lead companies down the wrong path to a series of other pitfalls and tricks.

"Instead, view SOA as a broad set of design concepts centering on your major business processes and transactions,  and view Web services as one set of application-to-application communication protocols by which to access your services," he explains.

Another idea that follows that one is the idea that SOA is a technology only. That's a very small view of SOA, says Heffner, and what's more important is the idea of SOA as a business design to enable strategic transformation and flexibility to optimize processes, which is the higher level view of SOA.

And you don't necessarily need to buy loads of new products that break the bank in order to do SOA, he continues. "While it's true that you may eventually buy enterprise services buses and SOA management and repositories and appliances and such, you may well be able to get started on SOA and achieve strong business benefits without buying anything new."

SOA Truths (or Treats)

No one wants tricks, so how about some treats? Heffner says there are plenty of them in SOA. "The real treat within an SOA registry repository solution is not the registry, but the workflow and the service lifecycle tools around the repository," he states. That's what gets you a start on what makes SOA successful -- strong governance and organizational maturity to use a repository in the right way.

Considering SOA and Web services management, the real treat is visibility into the service implementation layer, says Heffner. A management product that looks only at requests and responses leaves you in the lurch if you need to troubleshoot something at the Java component or .Net component layer.

In an enterprise service bus, the treats may vary by the product but nearly all of them give you a unified access point for your services, he says, which gives you control over how services are accessed and how routing and various specifications are handled. And SOA testing tools give you strong and repeatable testing -- crucial as your company evolved and upgrades.

Heffner's Upcoming Keynote

Sound interesting? Heffner will be discussing all this and more in greater depth at his upcoming keynote.

"What we'll talk about in that presentation is focusing on the strategic view of your SOA platform," he says. "So, in some sense, starting with a more theoretical view in, 'Well, what are we aiming at? What do we want to get to with our strategic SOA platform?'"

A problem many companies encounter is how to get there while juggling different SOA projects on conflicting application platforms with different requirements and priorities for the infrastructure. And in order to put all that together, companies need a strong model for the SOA platform.

"You have to move towards that model by leveraging each project as a tactical step in a longer-term evolution towards that strategic platform," he says. "So in that session, I'll give the audience an overall view of what a strategic SOA platform is, what its major functions are, how some of the product categories of SOA infrastructure are playing out."

So in order to learn more about how to battle your company's SOA monsters, be sure to check back in to the SOA in Action Virtual Conference on October 31.

Executive Summary by Krissi Danielsson

Posted by krissidanielsson in SOA | Permalink | Comments (0) | TrackBacks (0)

Listen to the entire 9:53 podcast Download file

Posted by krissidanielsson in SOA | Permalink | Comments (0) | TrackBacks (0)

Listen to the entire 9:53 podcast Download file 

Don't miss Wednesday's 'A Two-Step Process That Will Simplify Software Delivery' Webinar with Skyway Expert Mike Evans!

In addition to the Visual Workspace tool itself, Skyway offers a set of processes that go around it, says Rodriguez. These processes “enable an organization to be more accurate in what it delivers, so as to ensure that the solutions that are built really match up to what business is looking for,” he said. “That process, combined with our workspace, enables it to be done much, much faster than can be coded today.

What's New in Version 5.0

Rodriguez feels that a disconnect between customers and IT is one of the biggest problems in the software development lifecycle today.

“The way that a customer is queried upfront and use cases are generated and these long requirements-gathering processes go on, while often very handy in trying to understand what the customer wants, never really fully addresses the problem,” he said.

Rodriguez cites estimates from analysts that as many as 80 percent of solutions delivered may not match with what businesses want, so the latest iteration of Skyway Visual Workspace gives customers the ability to have an interactive demo to see what they’re getting in advance.

“The prototype is 100 percent leveraged into the code and the models that end up being deployed long-term,” he said.

Version 5.0 of the software also includes a model-driven AJAX environment inside of Skyway that enables the building of applications with graphical models. This means users can build rich user interfaces that look at act like rich clients, but without a lot of XML and JavaScript.

The new version also includes the ability to integrate with Adobe DreamWeaver and added support for distributed transaction models.

Comparison to Typical SOA Offerings

Although many SOA offerings revolve around the runtime side of things, Rodriguez points out that Skyway plays in a different space, focusing on delivery and deployment.

“Skyway tools are really all about how do you build these things and then deploy them out into the environment and not about how do they run after the fact,” he explains. “What comes out of the Skyway Skyway platform is industry-standard J2EE code that's all packaged up and ready to deploy to an application server and enabled as Web services."

Compatibility with Legacy Applications

What if a company has an antiquated ERP or CRM system? Rodriguez feels that Skyway has great offerings for how to extend legacy systems without actually having to replace them. He cites the example of Southern States, which had a perfectly functioning JD Edwards system that worked for every area except customer pricing. The Skyway Workspace helped Southern States to extend their current system but to build out new logic and new UIs to manage custom pricing.

“That's actually a common thread,” he said. “We see that in most of our customers. They're not building Greenfield applications; they're extending what they've got that's inside their enterprise."

To find out much more about the topics summarized above, including which Skyway customers are already using the 5.0 release, listen to the entire 9:54 podcast.

--executive summary by Krissi Danielsson

Posted by krissidanielsson in SOA | Permalink | Comments (0) | TrackBacks (0)

Listen to the entire 17:40 podcast Download file

It's been a busy week over at Vitria Technology Inc., with not just one, but two interesting product releases that will enable faster and better SOA development.

Business Accelerator is an open, agnostic enterprise-class integration suite built from the ground up for Service-Oriented Architecture (SOA) and Event-Driven Architecture (EDA). Resolution Accelerator is designed to help process business exceptions and improve business processes in general. It seems to have, and -- the pun is intended here -- some exceptional capacities to ferret out and repair exceptions’ adverse effect across an entire enterprise.

“When we went out and surveyed customers, we saw that there were a few things that they were looking for in SOA technology,” Dr. Skeen said. These included portal-building ability, new was to incorporate business processes, business process management and integration-centric technology and the ability to leverage what they had done so far.

“So many times they go to vendors and they find that if they want to buy a component, they have to buy the entire SOA stack,” Dr. Skeen said, adding that many clients were caught between from what they saw two different types of tools. “One allowed to implement quickly and deploy quickly but maybe not be able to, if the project was successful, scale up to an enterprise scale,” Dr. Skeen added. “Other tools had the proven enterprise salability but were often hard to deploy to begin with, and again often came with proprietary SOA stacks,” he added.

Finding the Sweet Spot -- and Some Twists

To find the sweet spot between these desires, Vitria took decided to stay with its historical focus on providing strong integration technology and 12-year history building integration BPM tools.

“But we also decided to bring this technology to market with a twist,” Dr. Skeen said.

--Allow their technology to be used with other strong ecosystem partners that they have certified against.

--Make it agnostic by allowing it to be “once-click” deployment on top of other SOA platforms, specifically app servers and ESBs.

“This also allowed our customers to go to market quickly with current technology they have in place, even though they know over time they may have to evolve this technology,” Dr. Skeen notes.

“Also, Gartner, for example estimates, that typical SOA enterprise experiment with SOA will not have one SOA environment but anywhere from three to more, and these environments typically have to be maintained separately,” he noted.

“So, for example, if you have a JBoss environment in one geographic region, maybe a WebSphere environment in another geographic region and maybe a WebLogic environment in yet another region, still, instead of having developed three different integration solutions and deploy them into three different environments, what we allow you to develop it once and deploy it in those three environments,” Skeen said.

Vitria’s strong ecosystem of partners includes RedHat, for app server and middleware messaging services, AmberPoint for the management of Web services. Skeen noted that Vitria “was also working with IBM and BEA, companies which we have viewed traditionally been our competitors. But we think, in this new world, really we need to all be viewed as partners.”

Exceptional Exception Handling for Increased SOA

Skeen then turned to Vitria’s second upgraded product, Resolution Accelerator 3.0, which is designed to quickly and cost-effectively handle exceptions.

“Unfortunately, for many companies, exceptions are not that exceptional, i.e., they occur more frequently than what one would like,” said Skeen, who went on  to detail how they occur at all level of an IT stack, including:

• System exceptions, such as a database or an application going down.
•   Higher-level exceptions such as a Web service is not responding in time.

“Or even a higher level, a process for example, you submit an order. And that order needs to go through a number of steps. You need to check customer information, billing information, for example. Inventory -- do you have that in stock?” Skeen noted. “All those are likely to be different systems and any point in that long process, something could fail.”

And he added that SOA’s decentralization and multiple technology could even exacerbate the problem by making exceptions harder to locate or event.

Skeen describe how Resolution Accelerator’s solution is to uses the concept of “mass repair” to tap into all levels of exceptions by handling them it at a business-problem level and restarting all the processes that were involved in the effects of that root cause.

“So, for example, if your order stops, then what you really want to be able to do is find some way to continue that order, as quickly as possible, so you don't lose that customer,” noted Skeen added.

Mass Repair in Action

“We find that a lot of times, many problems that may occur, for example, orders being stopped, may be the result of a single problem happening at a lower level -- for example, an application goes down,” Skeen said.

“Or, if you're an energy company or a power company, your first sign of a transmission line going down may be a thousand customer calls coming in. Now clearly, those thousand calls, you need to be able to take care of, but there is one underlying cause, and once you've identified that root cause, you'd like to fix it and then be able to make sure that those customers can be taken of, or in the case of an application going down and orders being stopped, those orders can progress nicely,” Skeen said.

An added bonus is that Resolution Accelerator runs on top of Business Accelerator’s open and agnostic architecture, “which means that means that you can go into an environment with Resolution Accelerator and leverage your current SOA infrastructure. You don't have to rip and replace, you don't have to install a new stack. You can really leverage what's there,” Skeen noted.

Dr. Skeen also decribed how early adopters like telcos and financial services companies at a mature stage of SOA can use the compbination to solve key problems, such as automating their enterprise business processes by providing better insight through their governance and Web services tools into the effects of many exceptions.

“So we can really leverage the tools that really come into play and take it from just managing exceptions at the service level to really managing the exceptions at a business level,” he noted.

“In the agnostic side as well, these companies can take advantage of the fact that they probably do have multiple SOA environments, and they would like to rationalize them, perhaps reduce them to one or two,” Skeen said.

Other industries, such as healthcare services that are just at the beginning of SOA adoption “will get a complete ecosystem of partners that can out of the box provide a total SOA solution,” Skeen said.

“They can choose just to use the technology out of the box and focus only on the problem they have … later on if they need to make a different choice, they are still free to do so.”

To find out much more about the topics summarized above, listen to the entire 17:40 podcast.

Posted by gtrotta in SOA | Permalink | Comments (0) | TrackBacks (0)

IBM’s Worldwide Product Marketing Manager for WebSphere Application Integration Leif Davidsen sees five entry points that enable the IT infrastructure changes that lets organizations embark on the road to SOA.

“Three of them are business focused – people, process and information and how the business might reflect those parts of itself in SOA,” Davidsen says. “And then we have two more technical entry points, reuse and connectivity.

And the drawbacks of building point-to-point hard coupled links between and more and more interface logic into the applications themselves includes:

--Swamping  the actual business logic that the application is actually there to perform and increasing the cost of changing or maintaining that application.

-- Introducing the potential for errors or fragility.

--Rendering the applications almost completely unusable in an SOA environment.

“You want to then reuse them as services, where services are actually trying to get to the core business logic itself rather than be interested in the specific way in which they are invoked in terms of an application interface,” Davidsen says.

Untangling the Rat's Nest

He went on to detail IBM’s three-tiered approach to letting an ESB provide a completely decoupled infrastructure allowing a business to make use of the actual core business logic where all interface layers were defined and managed and maintained within the ESB layer and not in the applications themselves.

At the simplest level, an ESB can meet a company’s specific business needs by leveraging the decoupled connectivity that IBM’s WebSphere MQ transport layer has provided between applications for more than 12 years.

“It really provides much more intelligence in terms of routing the information between applications and services and also transforming and performing intelligent processing based on the content of the flow, based on awareness of each application that’s actually both sending and receiving the information," Davidsen noted.

"So there’s all sorts of different ways in which you can use ESBs throughout your business today," Davidsen said, before detailing actual case studies by a wide range of companies.

“What’s always been encouraging has been that these customers span both the globe in terms of where they’re running their businesses and in terms of the different industries that they exist and successfully perform in,” Davidsen said. They include:

--HypoVereinsbank:  “Their line of business, they were finding, they were stuck in a silo mentality in terms of only wanting to connect to their own applications, which obviously restricts their growth capability," Davidsen said. "So they wanted to implement an ESB to allow them to move to a much more component-based building block model for their business, allowing their services to be more widely used across their business.

--Auto Leaser Ubench: “They’re using very innovative technology -- wireless telemetry devices -- in their leased auto fleets that allow them to offer capabilities such as automatically detecting when services or maintenance is required and sending that back to the other value chain suppliers across their leasing business to arrange that," Davidsen said.

“And then there are other companies, like a media company, VRT, who wanted to put in place an ESB to connect their digital and tape-based media assets for both faster, efficient storage and retrieval of those assets," Davidsen said.

VRT, Davidsen noted, “is actually using one of our other ESB products, WebSphere ESB, that’s ideally designed for service-based integration. They had built an almost completely Web services based infrastructure and the data types, the metadata, were all easily describable in terms of XML and sort of more open data structures, so they could use WebSphere ESB to provide the intermediation between the different applications and data types.

DataPower: An Advanced Solution

Another company, digital rights manager Macrovision, was running WebSphere ESB at the heart of their process server and had an e-commerce front end to their business.

“We believe that it will be much more difficult to really meet our customer needs by building a single ESB that tried to do everything, because those customers that needed a simpler product would probably be overwhelmed,” Davidsen said. “And it’s like if you tried to cut these things back too much, you’re going lose some of the capability and flexibility and scalability that many of our really large enterprise customers find that they need.

“What we find in terms of the implementation model is that businesses would typically implement a particular ESB from our portfolio for a particular departmental or project needs as part of an overall ESB strategy, in the full knowledge that any combination of these ESBs would work together. So you can have WebSphere ESB connected to WebSphere Message Broker, or DataPower connecting to either of those two,” Davidsen said.

For much more on the topic – and more case studies – listen to the full podcast.

Posted by gtrotta in SOA | Permalink | Comments (4) | TrackBacks (0)

Web 2.0 Data mashups are to SOA what NASCAR is to automobile production -- a much faster, more free-flowing, results-oriented way of combining complementary components into new applications that have an advantage over traditional months-long application development or production cycles.

And they seem to be rivaling NASCAR in popularity as large, medium and small software companies are all offering mashups. At the forefront of the movement is San Francisco-based software company ActiveGrid, which is offering free on-site "Build Days" to qualifying companies to deploy their mashups.

“The key advantage is the speed to which business-driving applications can be delivered to the end user base within the organization,” said Todd Hay, ActiveGrid’s Vice President of Marketing and Business Development. “Their customers and their salespeople can get the information they need immediately instead of waiting for the long term, long-run enterprise integration technologies to finish.”

Hay went on to describe how ActiveGrid:

--Leverages a firm’s existing techniques to ensure security, reliability and scalability while “still giving them the flexible, easy-to-use and fast-to-deliver front ends the Web 2.0 or mashup world has driven.”

--Provides the tools to allow developers to quickly build applications out of the services and out of capabilities being delivered by giants like IBM, Microsoft and eBay.

-- Disptaches their top engineers to a company to show them how ActiveGrid “can very quickly solve their business problems on their turf with their technologies. And that has been a very effective approach for us.”

--Enabled one manufacturing giant in the Midwest to pull marketing information out of a number of their field databases and create a marketing analytics application that allows them to create corporate-wide marketing campaigns across their global presence of retail manufacturing outlets.

For mashups, rapid delivery, rapid iteration and extendibility according to end users’ needs are the key requirements, Hay noted.

“I do see mashups as the logical evolution of where Web services were trying to go and where the Web was trying to go; service oriented architectures in general have been laying the groundwork for this for years,” Hay said. “We’ve now just advanced the technologies to the point where it is delivering on those promises.

Hay added his thoughts on the ideal, complimentary balance between an SOA and mashups:

“A large customer or ours was very excited that ActiveGrid can come in and start building Web 2.0 AJAX-enabled mashups based on their existing databases and some external Web Services. And as they build out a large SOA infrastructure from the back end with the governance and repositories they’re talking about; they can migrate and leverage those applications moving forward.”

For many more details, listen to the entire 7:29 podcast Download file

Posted by gtrotta in SOA | Permalink | Comments (0) | TrackBacks (0)

Over the years, I’ve found that I’m listed in some data repositories as Gian Trotta, Gian G. Trotta, Trotta, Gian and as John Trotta and Gina Trotta to those who have opted to anglicize or feminize my name.

And that’s just one example of what makes it hard to integrate data from different databases in SOAs. But last week, Progress Software unveiled its DataXtend Semantic Integrator – and its model, mapping and schema creation-tools are potent aids to validating and transforming data access across SOA applications.

“Whether it’s a relational database, it’s in a Web service or it’s in a flat file,” noted Progress Software’s Vice President for Data Services Ken Rugg. “It’s very important that if these systems are going to cooperate, that they’re treating you as the same person every time you go to their systems,” said .

The next challenge is to go beyond format integrity and assure consistency with business requirements.
Data quality and data consistency is a big theme to both service customers appropriately, ensure regulatory compliance.

“The way we see it is is that there are number of tools that give you the ability to maintain the data consistency or cleansing the data,” Rugg added. “But with the DataXtend integator product, we’re dealing with data in motion that’s flowing between services and from one data source to another. And in those cases we present a way of creating rules that insure the constraints and consistency of the data is managed throughout its lifecycle while it is in motion, while you’re able to use it.

Rugg also detailed the advantages of allowing customers to both create and map between custom schemas and industry-standard ones.

“One of the largest scenarios where we see our customers using the technology is when they’re using an industry-standard schema, something like an HL7 or a SID from the telecommunications space,” he noted.

They have these very large models; they want to be able to have an informational model; a canonical model that represents their entire business," Rugg added. "For a significant enterprise, those models can be very large and very complex and they can be hard to manage over their lifecycle because they do certainly have to change to adapt to the business. When we talk to our clients, being able to handle that complexity and that ability to handle those large-scale schemas is really one the key things they’re finding useful in our tool."

An early case study included a major telecom provider using DataXtend to integrate their operational and business systems. “It may be information about you as a customer from their landline support, their cellular support and their broadband support and they want to make sure it’s Gian that they’re talking to, no matter what it is they’re talking to you about, Rugg noted.

“You might see upwards of 3000 individual point systems that are being use to bring this common data and information together. That’s a challenge that really requires a tool of the class and quality of DataXtend’s capabilities,” he concluded.

For many more details, listen to the entire 6:49 podcast Download file

Posted by gtrotta in SOA | Permalink | Comments (0) | TrackBacks (0)

Are you one of the many companies that have built SOA applications that no one is using? Last month, NetManage announced SOA Planner, a new Java-Based planning tool that lets organizations create SOA services based on actual usage patterns.

It also enables what Archie Roboostoff, senior manager of product management at NetManage calls "Incremental SOA."

It’s a strategy for creating services around legacy-based assets that allows customers to incrementally plan, build, evolve and scale a service.

“Many business functions cross legacy applications, and this is something SOA Planner can expose,” Roboostoff notes. “For example, a customer lookup function might go four screens into a transaction, and an end user may copy content onto three screens on another system and then come back to that original system and complete that transaction.”

As a result, a non-developer can use SOA Planner with NetManage's OnWeb tool to export a script that will enable a service.

“So quite literally, by having my end users do their everyday business, I can literarily have them create their own services and I can deploy those to OnWeb,” Roboostoff said.

"This is something that is not available today and is resonating quickly throughout our industry," Roboostoff added. "While this is a new product, it still stays in line with our mission all along, when it comes to connectivity: a customer should not have to be an expert in the systems they’re integration with, and they should be able to deliver via any technology and any application server and any framework."

Roboostoff went on to detail a case study of a large financial institution that wanted to get a top-down look at how their users were using their legacy assets and legacy information.

“Using SOA planner, they’re able now to from their central location, track and tag specific business units, and now how they have quantifiable evidence that contradicted some of their early thoughts,” Roboostoff said.

“Now they have that quantifiable evidence as to what should be service-enabled and why and they can get this off the whiteboard for the first time and into a prototype environment so they can the impact of these services on everyday business,” he added.

For complete details, listen to the podcast (4:49 length) Download file

Posted by gtrotta in SOA | Permalink | Comments (0) | TrackBacks (0)

A recent ebizQ survey showed organizations facing some formidable problems with SOA governance -- but some automated help is on the way.

SOA Software today announced Workbench, a comprehensive standalone, closed-loop SOA governance product that could simplify governance and speed service adoption and reuse.

“Governance is really about instilling consistency and accountability and encouraging desired behavior,” said Ian Goldsmith, SOA Software’s vice president of product marketing.

“You can’t just lay down a governance solution which set all sorts of onerous requirements and makes it difficult for people to begin to leverage the tools in place. You have to make sure that you make it easy -- and that’s what Workbench is all about,” he added.

Goldsmith went on to detail how deploying Workbench with SOA Software’s Service Manager product provides closed-loop governance.

“The alternative is distinct, separate products using a sort of ‘define and hope’ model of governance from a registry repository where you set up a policy and hope or pray that something’s out there randomly enforcing it,” he noted.

“And then the policy enforcement products, the run-time platforms end up in what I term a ‘fire-ready-aim’ model of policy enforcement where they simply randomly enforce things and no one really knows what it is they’re enforcing,” Goldsmith added.

The ebizQ governance survey also found that confidence in governance grew in direct proportion to its degree of automation.

“It’s very encouraging that the survey found that correlation because it gels very nicely with what we’ve seen and I think it also comes back to reflecting on the idea of ease of use being critical,” Goldsmith said.

Goldsmith described Workbench’s automated workflows, such as having Service Manager automatically detect a service in deployment and submit that process to a workflow process in Workbench. Workbench can also marshal a new service through a series of policy compliance checks for design-time compliance through a series of approvals to insure only appropriate services are published in the organization in the first place.

A second workflow can define “what we call and active contract -- an XML document that defines the contractual relationship between a consumer and a provider,” Goldsmith added. “So it specifies the policies the consumer has to the consumer has to comply with, it specifies the SLAs that will be monitored at both ends, it specifies the capabilities of the provider, it specifies the times the consumers can access it, it specifies the capacity they’re allowed to access and whole bunch of things; it can even specify mediation between incompatible standards and technologies.”

Goldsmith also described how Workbench can help strike a balance between controlling the proliferation of services and allowing non-developers to create services.

“For example, one of the things it allows you to do -- and this is really very innovative -- it allows a potential consumer of a service to specify their own service definition and submit that service definition to development process for execution,” Goldstein noted.

“So the development side can look at these submission an see which ones they want to submit and which ones they don’t and whether they want to enter into a formal contract with a consumer before they develop it,” Goldstein said.

For much more on Workbench -- and SOA Software’s explosive growth in the last two years -- listen to the entire podcast.

Download file

Posted by gtrotta in SOA | Permalink | Comments (0) | TrackBacks (0)

I’m Beth Gold-Bernstein, Director of the ebizQ Training Center and Chair of our SOA in Action Conference – and I’d like to detail a few of the conference’s key themes and speakers.

On November 16th and 17th, we will be presenting a live virtual tradeshow that will include Webinars and online booths in an interactive environment which allows you to chat with leading experts, vendors and attendees. It’s all right here, and it’s free.

We have an exciting agenda lined up for you. On Wed Nov 15th at 10:30 am ET Roy Schulte, VP Distinguished Gartner Analyst is going to give a keynote on “Meeting the Challenges of SOA Adoption.”

On Thursday, November 16 at 10:30 am ET, Ken Vollmer, principal analyst at Forrester, will give a keynote on “The Relationship Between BPM and SOA. “

Each day we also have an exciting line-up of other Webinars on planning, building and managing SOA and panel discussions on Understaind ESBs and SOA Governance.

In between sessions you can visit the on-line booths, win prizes, interact with vendors and other attendees, and download literature.

You can also visit the ebizQ SOA in Action Resource Center, the premier online destination for pragmatic and actionable information on SOA , where you'll find Webinars, podcasts, blogs, articles, polls and analyst reports on planning, building and managing SOA solutions. It’s available now, and it’s free.

Thanks for attending our virtual tradeshow. Please let us know what you think, and if you like it, be sure to invite your friends and colleagues.

Listen to Beth's entire greeting. Download file

Posted by gtrotta in SOA | Permalink | Comments (0) | TrackBacks (0)

TIBCO on October 16th released Collaborative Information Manager (CIM) 6.0, a central Master Data Management (MDM) repository that aligns data so that business services and composite applications within an SOA have accurate, consistent and timely information.

“The challenge in master data management stems from keeping the data current making it stay current throughout the life of the data changes,” said Neeraj Gokhale, General Manager of TIBCO’s CIM Business Unit.

Gokhale described how CIM’s an event-based architecture automatically updates data changes with central depositories "from data conception to data consumption."

"When data does gets changed based business rules… we provide the governance and policy management to publish those changes selectively to the downstream systems that need to know about these changes,” he said.

“What you can do is create a complete alignment in master data from point of change to point of consumption of the data where the transaction systems, such as perhaps a an ERP or CRM system can be informed of the changes,” he said.

Gokhale also detailed a variety of case studies, including a Fortune 500 pharmaceutical company that integrated a wide range of enterprise software in the wake of numerous acquisitions.

Another benefit is the ability to both structured and unstructured data.

“Retailers are inclined to use a variety of rich product information so they can power their Web sites through multiple new channels,” Gokhale said. “Similarly financial services are maintaining for a variety of compliance data behind the securities that they trade.”

He cited the example of a consumer manufacturing company’s need to provide up-to-date product information sheets.

“CIM not only stores the rich elements like pictures and bar codes in synch, but creates that .pdf as an output at the end of new product introduction process,” Gokhale noted.

“We can assimilate all the up-to-date and accurate product information that is made into the form with a scannable bar code and picture of the product that can be e-mailed as part of a business process to all the executives in charge of that product, so when they need to provide that information to their customers they have an accurate data representation in paper format,” he explained.

For complete details, listen to the podcast. Download file

Posted by gtrotta in SOA | Permalink | Comments (0) | TrackBacks (0)

Here's some news for companies building an SOA – there’s a new approach to coordinating business and IT groups and ensuring a higher level of quality, change, and build management to deliver services.

Dave Schwartz, General Manager of Mergere, joined us to explain. His company’s new 1.1 version of its Maestro product allows enterprises to keep a well-organized repository of what services they are using and a transparent view into how they are being used.

“The service-based IT infrastructure and good build tools are driven by three key concepts,” Schwartz said. They are, in his words:

1. Broad manageability of interdependent components across an asynchronous, heterogenous environment.

2. Good separation of concerns by using modularity and granularity in the development process.

3. Standardization of interfaces and dependencies.

Of the three, “broad manageability is definitely the key thing that everyone’s focused on right now,” Schwartz said. “And that begins with the active use of an articfact repository that contains artifacts you’re going to need during build time and run time and also metadata about these types of information.

“Organizations that maintain this information early in the development process are guaranteed that they’re going to be able to have manageability once they deploy the service in the long run,” Schwartz said. “This type of information can automate the actual building of the service and make development of services much simpler for SOA developers.”

On reuse, “you need to know a component exists and you also need to know how to get that component,” Schwartz said. Having an artifact repository available and being able to do continuous integration testing against these types of systems ensures you can have reuse within your organization.

If you don’t have ability to test the service by deploying the service automatically in an integration-type environment, you’re not going to know if your interfaces are clean and will build the way you’re going to need them to at run time," he added.

Schwartz concluded by explaining how Maestro tracks actual cross-dependencies between the different applications in metadata. “When you know this from the beginning there’s no additional info you’re going to need to acquire to provide the governance,” he noted. “You can use our repository to see all the different applications that are making use of our service and let people proactively know about the changes.”

For complete details, listen to the podcast. Download file

Posted by gtrotta in SOA | Permalink | Comments (0) | TrackBacks (0)