The ITKO Blog: Service Virtualization and Application Lifecycle Optimization - CA Technologies

SOA Success Stories Part Four: Global High-Tech Manufacturer

Here's another tale in our series of field stories to show successful cases of ensuring quality and agility for dynamic enterprise apps. This case covers how a global high-tech manufacturer tested and delivered a demanding enterprise e-commerce site under time and budget.

In this case there was a lack of visibility and traceability that was posing a serious threat to mission critical business operations. Unlike some of our other cases, this company had complete control over the design and implementation of this complex multi-tier app, yet they still had no way to trace a single action to a single source in the maze of multiple service layers.

Ruston Vickers, iTKO's CTO & co-founder, worked on this effort. He said that before the test automation project, the company had to do manual sequential tests on each subcomponent to find error sources: "It was like having to place a plumber at every faucet to find the source of stoppage in a house with hundreds of rooms. Instead of having the plumbers yell back and forth across the large house as they progressed through the system, turning on faucets one step at a time, LISA gave the company an x-ray machine to see a top-down view of the entire system at once and determine the cause and effect."

This top-down viewing capability cut their best prior testing time in half. Formerly the company employed 10-15 consultants just to continually test only the end result UI behaviors of their system. Now, with a highly reusable and complete test suite in place through iTKO LISA, the consulting firm can conduct the same process with one part-time resource, at a higher level of accuracy, as the automation capabilities of LISA erase the margin for error inherent in manual testing.

The company is pleased with the reduced costs and increased speed of implementation, and the consultant on the project is pleased with the increased value they can provide to their client. With LISA they both have a reusable asset that will continue to provide value as they go forward together with new releases.

Reusable, continuous validation activities save hard dollars in testing labor costs, and the number of support costs incurred, but cost savings aren't the only value. In a customer-facing environment, eliminating 1-2 months of time to successful release can make a huge impact on incoming revenue potential.

Stories from the Field - Part Three: Major European Telco Company

This is the third installment of an open-ended series of enterprise quality & virtualization success stories from our field engineers. It covers how a major European Telco company ensured BPM quality with test validation across all application layers.

"This was a classic case of being caught in the middle with moving targets all around," reports Ruston Vickers, iTKO's CTO. The telco needed to be able to validate downstream B2B integration after taking an upstream provisioning request, processing it as the middleman, and passing it on. The problem was that they had no control over the upstream and downstream services. "Handling this validation issue by calling on the real up and downstream processes was not feasible, and mocking them up manually required too much work and left you trying to model and hit simultaneously moving targets," Ruston said. This made it almost impossible to track down the source of many bugs.

Ruston's team solved this mission impossible problem by providing the capability to accurately simulate both ends, the middle, and the entire process in real time. LISA can first act as the upstream requester, launching the process and then it following it through the multiple service layers to completion. This is done through invocation, validation, and where services are designed but not yet built, the addition of Service Virtualization to simulate the needed services to complete validation.

The actual live upstream and downstream services are then no longer required for testing and validation, freeing the middle level services to rapidly test independently, also avoiding any third party transaction costs or data management issues. The result of implementing iTKO LISA's Validate and Virtualize products was transformational for this telco, and included the functional validation of 120 different order types, as well as load testing of TIBCO BusinessWorks processes and database stored procedures.

Higher quality was achieved through near 100% test coverage to validate end-to-end business process outcomes. In addition, there was faster time-to-market by removing system availability constraints and reduced transaction costs. You can read the complete case here. Our next case covers how a global high-tech manufacturer tested and delivered a demanding enterprise e-commerce site under time and budget.

iTKO & Cognizant Talk about Software AG Quality - Webinar

We're having an iTKO-hosted webinar Wednesday on a topic we revisit quite a bit, ensuring quality and agility for Software AG implementations. Our Software AG customers have very unique needs - from building and validating next-generation SOA and BPM architectures, to regression testing and virtualizing test labs while conducting an upgrade of their infrastructure to the latest version of webMethods.

iTKO's own Founder & Chief Geek John Michelsen will be joined by an expert from one of our key SI partners from Cognizant, Alfredo P. Reyes, the EAI Practice Head - North America, who will talk about best practice methodologies. This should lead to a very good discussion, so bring your questions.

Registration is free here:
"iTKO Webinar: Reducing SoftwareAG Implementation Costs and Risks with Collaborative Quality Across the Lifecycle"
Wednesday, February 25, 2:00PM Eastern

According to the webinar invitation, "Enterprise software implementations today are up against extremely difficult cost and time constraints. The need for greater cost savings, risk mitigation, quality, and end-to-end control for SoftwareAG/webMethods implementations is driving more customers to evaluate new solutions to optimize their software testing and delivery lifecycle."

Stories from the Field - Part Two: Leading Financial Institution

This is the second of an eight part series on "stories from the field" from our engineers and architects working with customers to solve the challenges of dynamic enterprise applications, based on SOA, BPM, integration suites, and ESBs.

It looks at how a leading financial institution eliminated a cumbersome and expensive manual testing process, freeing up resources and dramatically decreasing implementation cycle time. The bank set up a center of excellence (CoE) for middleware performance testing. In many cases one middleware test might connect with 70 to 80 systems that represent an alphabet soup of different technologies. To handle this effort, the bank built a very large custom test harness to respond to their load testing solution over the course of several years. But it could not meet all the changing needs and required a team of consultants just to keep it going.

The hand-coded responders were all one-offs that were not integrated, requiring so much rework that the team needed to push back releases by months to accommodate test cycles. Ken, the iTKO architect brought in, said that the bank felt this was not a scalable or cost effective solution anymore. Ken was able to "virtualize" or simulate these 70+ applications running within a Virtual Service Environment (VSE) in a few days - with more dynamic logic than the hand-built responders that had taken years to put in place.

Now they had an integrated solution that could handle the entire process through an automated testing approach that could provide ready access and data modeled to adapt to needed test scenario changes. The bank had been employing a team of highly skilled and highly paid consultants to maintain the test harness and perform testing – much of it manually. These consultants had been originally brought in for more complex application and test case development. Now they could focus on their original assignment. It was a win all around.

The bank greatly reduced their ongoing testing and validation costs and time. Several additional product releases per year are now planned. The solution consulting partners were also winners, as they were able to return staff to higher value tasks, maintaining a service level differentiator over lower skilled competitors for the manual tasks they had been pitch-hitting on.

In fact, the consulting company was so pleased by the outcome that they have started to build a consulting practice around the use of iTKO LISA for similar situations at other financial service institutions. It's informally been called the "bank in a box" solution because of its power and simplicity. We all know how much financial services firms need to cut costs these days so they should find a responsive audience. Go here to see the entire case. Our next case covers how a leading Canadian Bank removed constraints to greater productivity in the Test Lab.

The Cloud Will Grow in 2009 - IDC

What IT sector goes up in an economic down turn? IDC answers the question with this report, Software as a Service Market Will Expand Rather than Contract Despite the Economic Crisis, IDC Finds.

Cloud Computing would have grown in a good economic environment - but this continued growth is a sign of the times. The cloud is good news for companies seeking faster ways to turn on revenue-generating functionality and save up-front costs.

Highlights of the report include:
1. By the end of 2009, 76% of U.S. organizations will have at least one SaaS-delivered application.
2. The percentage of U.S. firms that plan to spend at least 25% of their IT budgets on SaaS applications will increase from 23% in 2008 to nearly 45% in 2010.
3. This market's growth prospects will accelerate the shift to SaaS for the whole value chain as the promise of a recurring revenue stream, and the opportunity to tap OPEX and project-related dollars, will benefit the whole SaaS ecosystem.

These projections for the growth of the cloud are also consistent with our experiences with SaaS’s cousin, SOA. The cloud won't make complexity and risk go away, even if it seems to obscure much of the initial setup effort for end customers.

As companies increasingly rely on the cloud to support business processes, the interactions shared among multiple new services will need to be tested and validated. Per-use service costs and availability will also create significant constraints, and the need for service virtualization will become evident as a means of getting the best of the cloud while eliminating cloud dependencies.

SOA has been around longer than SaaS and is running ahead of it on the Gartner hype curve (See Will SOA Fold into SaaS? Will Quality Happen Along the Way? As I wrote then, one obstacle for the pure plays in the cloud to join SOA in the Gartner slope of enlightenment is to recognize and deal with the increased testing and validation of data from even more diverse sources, as well as the governance and validation of all the underlying technologies that contribute to specific business outcomes.

Field Reports on SOA Success - Part One: National Cable Provider

Recently, there has been a lot of fun discussion about the death of SOA. Well SOA successes are quietly happening every day, and one of the keys is the use of robust testing and validation and, some cases, the use of service virtualization to speed up the testing and validation process.

SOA's shine may have worn off for some, but it is just another practice for designing and building dynamic, distributed enterprise apps. SOA may share boundaries with other technology practices: BPM, integration suites, and ESBs. All of these describe systems that are heterogeneous, distributed, multi-tier, and subject to frequent change by nature.

So call it what you will, not meeting the new challenges raised by this class of apps can lead to various unintended consequences and failures, or create bottlenecks to agility and cost savings which were the ultimate goal. We want to add to the conversation with a series of actual examples from the field to show successful cases of dynamic enterprise apps. This first one is a national cable provider.

This company had a very complex provisioning system - literally the "brains" that enables or disables what programming subscribers can see on their TV, depending upon requested configuration, on-demand status and special offers. The application was a complex application with several components that needed to interact at runtime in a secure transaction environment, and they had to operate efficiently and accurately to maintain customer satisfaction.

Our iTKO architect involved in this effort, Ken, said that before he and our integration partners at pureIntegration arrived, the cable provider staffers were hand building manual tests to try to cover these complex processes. Even a simple change could take the staff three weeks to build the less than adequate manual tests. Just getting ready to attempt to test a major release could take them several months, so the cable company was limited to only two major releases a year. This put them at a competitive disadvantage and added unnecessary expense to the testing and validation process.

Moreover, the QA manager noticed that manually testing was like trying to do deep sea exploration from a glass bottom boat. Gaining end-to-end test coverage and control, with automation, enabled them to move out of their testing silos and see the entire process. Now they can quickly spot error sources and correct them.

Ken said that in one case dozens of reported bugs were found to result from a single change. Getting past the 80%-20% rule and covering the most important and likely risk areas with automated testing first enabled them to spot these dependency failures in almost real time with each change. In the past, it was taking over three weeks to isolate the problem. Ken said that LISA was employed, enabling them to increase test automation and validation coverage from 15 – 20% to over 95%.

Ken added that “Saving several weeks in each test process has created a lot of excitement.” The results included the reduction of regression test execution from 3 weeks to 5-8 hours. The major release cycle was reduced by 6 weeks. The integration of LISA with the test and validation process also saved more than 95% of per-test costs, which is also an incentive to increase the test coverage to new features.

To read the entire case, go to iTKO LISA Case Study: Major Cable Service Provider.

Our next installment of this series will cover how a leading financial institution eliminated a cumbersome and expensive manual testing process, freeing up resources and dramatically decreasing implementation time.

Federated Computing demands SOA Governance & Testing

We recently read an article specific to the Federal Government's efforts that was very interesting and relevant. Our work with government agencies has exposed us to the concept of "Net-Centricity" which is a more overarching term than SOA. Net-Centricity (in my informal definition) represents a commitment of all departments and parties, across different authority domains, to work together to meet common goals through IT.

No matter how advanced the architecture design is, SOA alone won't deliver the vision of Net Centricity, with its expected agility, reuse and functionality benefits for the warfighter - if there is not trust instilled between teams and authority domains that must share and support an alphabet soup of IT systems and services. So in essence, the government's Net-Centric effort mirrors SOA and IT Governance efforts.

We recently read an article from Federal IT consultant Warren Suss that does a great job of covering this topic, titled "Streamlining IT Governance." It is good to hear someone talking about this. SOA Governance requires testing and certification of services, to ensure that there are no unintended consequences of change that can impact mission threads (or "workflows" in layman's terms).

In previous posts we have mentioned how Federal agencies are ahead of the business world in many aspects of SOA - due in no small part to this shared mission. In this space, we have seen these Net-Centric plans come a long way toward delivering a high level of quality and shared responsibility over the last 3-4 years.

Suss states: "Today, we recognize that important information can reside anywhere in the enterprise. What’s needed is a way to allow users with the need to know to get their hands on trusted information from anywhere in the network. Governance decisions, as much as technology decisions, determine the standards, rules, infrastructure and services that enable — or hinder -- the rapid discovery and sharing of information and capabilities."

We agree that Net-Centricity is about the practice first. Federal agencies seek to control a web of compartmentalized systems, by federating governance, so the teams that support the systems have some centralized standards, but not so much control that teams don't have local authority over local issues.

We must also talk about the technologies that support the federated environment of agencies and teams leveraging shared services. Strong testing and SOA policy validation on a continuous basis is crucial for ensuring trust in this type of shared environment. We also have a lot of research around Service Virtualization capabilities that eliminate the dependencies and constraints of these shared environments. When we can break interdependency and access constraints with virtual services, teams are able to develop and test without stepping on each others' toes in the software delivery lifecycle.

New Outsourced IT Paper - Collaborative Quality to Minimize Risk

Though we are in the testing and virtualization business, over the last 5 years we have had a lot of requests from customers for our advice about working with one or more outsourcing firms. Since we have encountered virtually every kind of working arrangement in our engagements, we do have some experience encountering both successful - and unsuccessful - teams comprised of multiple services partners.

Given the current market conditions for IT outsourcing, it seems like there is more uncertainty than ever in this realm. Outsourcing is still a must-do activity for at least some of any large enterprise IT budget.

For the most part we've kept our advice to the realm of working with our SI partners - as part of working with joint end customers, we need to cover how to better collaborate on quality, validate that business requirements like accuracy and performance benchmarks are met, and improve agility to deliver new features ahead of schedule.

There are many up-front aspects of properly selecting, negotiating contracts and defining requirements that will make or break a project - too much to cover here - but in our experience once the partners are selected, what defines the back-end of delivering a successful project?

The number one post-agreement issue we've found when companies have deployment issues and cost overruns in working with services firms, is that partners must be rewarded not just on delivery, but delivering with quality. We're not talking about simply handing down a set of requirements and enforcing them. Good collaboration is a two-way street - both the consumer and provider of services must be responsible for defining and delivering things that are tested, and testable.

Anyway, we wrote what we think is a pretty good primer on the topic - you can get the new paper here: "iTKO Whitepaper : Minimize IT Outsourcing Risk with Collaborative Quality" - http://www.itko.com/site/resources/collaborativequality.jsp. We can't tell you who to work with, but we can provide some advice we've seen on what works!

We'll be posting further thoughts on collaboration between multiple on-shore and offshore teams here - your thoughts and questions on the topic are welcome.

Can You Lend Me A Shovel So I Can Start My SOA?

Joe McKendrick (aka, Joe the SOA guy), put on his economist hat and recently commented on stimulus ideas for the downturned economy in his post, Shovel-ready SOA. He writes, "perhaps we should look at service orienting as a shovel-ready 'stimulus' program for corporate growth. Because it works both ways — the current rough-and-tumble economy may be good for SOA, and SOA may be good for the economy."

Good idea Joe. As we wrote earlier, on one hand SOA is not dead or going away. We see success stories quietly happening all the time. On the other hand, like the shovel, SOA isn't glamorous anymore, but it's being used to create real value. Not only that, but we recently were musing on how SOA virtualization can help cut carbon emissions while it is saving money, hitting another aspect of the current stimulus suggestions.

Joe takes this discussion of SOA and the economy a step further as he comments that it can also help in "revitalizing business growth as well ... Consider the advantages properly implemented SOA can deliver: Increase confidence through greater transparency of applications and data, especially at a time when markets have been quaking from fear of what they don’t know; help businesses collaborate better through more rapid integration… help organizations expand or contract pieces of the business, as needed, without fuss and muss."

In tough times, nothing boosts your chances to survive, and be positioned to thrive, like enhanced confidence, collaboration, and agility through SOA efforts. We'll just have to roll up our sleeves.

Of course, before you start digging, a little upfront architecture and planning goes a long way -- you don't want to hit a wire or gas line! OK, enough, we'll put away the shovel metaphor for now. Each of these themes can be further enabled by the confidence, speed, and agility that automated testing, ongoing validation and Service Virtualization can bring to SOA.