Ever try to integrate your most critical IT solutions? It's not for those with faint hearts—or falling budgets.
Fortunately, cross-solution integration recently got orders of magnitude easier, and more cost effective, with the introduction of the OSLC (Open Services for Lifecycle Collaboration) specifications. And in IBM’s Jazz for Service Management, an open architecture for integrating multi-vendor solutions, the strengths of OSLC are yielding tremendous results—results with impressive ROI, which we will explore in some detail in this article.
What's the case for OSLC and Jazz? Quite simply, they offer the end of a problematic paradigm: the homegrown, proprietary API-driven solution integration so commonly utilized today, and so easily broken by minor changes.
While custom code can certainly be created to provide an ad hoc integration between two IT solutions, that custom code typically presumes the solutions themselves will remain constant. And should one or more of them be upgraded, the integration stands an unfortunate chance of disintegrating.
OSLC (and Jazz) offers a different approach—one that multiplies the potential ROI from integration. To do this, OSLC leverages an open set of specifications, spanning solutions and solution vendors, that defines what kinds of data resources should be shared, and how that sharing will take place. The idea is to leverage the proven practices of the Web (links) to reference and interconnect data, and make any two IT solutions interoperable. Equally importantly, OSLC does so in a way that (because it's standardized) will be far less brittle than integrations based on custom coding.
Instead of breaking with every new point release of a solution, integrations will continue to work properly, rendering IT services more available, reliable, speeding problem resolution, and helping organizations avoid the problem of "vendor lock-in" that might otherwise apply. For instance, if organizations wish to deploy cross-vendor solutions that were never originally designed to be integrated, the desired integration is possible by exposing a minimum set of resource attributes based on the OSLC specification—this approach is often referred to as an OSLC façade or wrapper. If at some point in the future one or both of those solutions must be replaced for any reason, any vendor’s OSLC-compliant offering, or “OLSC-wrapped” offering, can simply plug in, and provide equivalent capabilities.
IBM's own implementations of OSLC, collectively known as Jazz, have already experienced considerable success in the IBM Rational software development portfolio. And in the last year, IBM has also applied these design principles to its Cloud and Smarter Infrastructure portfolio with impressive results. Already, Jazz for Service Management has been implemented in key offerings including IBM SmartCloud Monitoring, IBM Tivoli Monitoring, IBM Netcool OMNIbus, IBM Netcool Impact, and IBM Tivoli Business Service Manager—and more will be coming in the very near future.
As a result, the promise of Jazz—to improve business value in both qualitative and quantitative ways through simplified, accelerated, and resilient solution integration—is already being realized, and in many different kinds of deployments. Three scenarios of this promise being realized follow.
Scenario: Accelerating problem resolution, yet also reducing operational costs
One of the most compelling examples of how Jazz solutions can create value arises from one of the most common problems in IT operations: a complex technical issue that defies easy or rapid analysis.
Because Jazz supports cross-solution data links, it's possible to create a single, unified view of relevant, in-context data—the ideal resource for troubleshooting purposes. This helps to diminish both the time-to-resolve and cost for each such problem.
Many IT issues, for instance, can cascade in unpredictable ways—rapidly emerging from a root cause to affect multiple systems, applications, and services, yielding a chain reaction that is hard to identify or understand. A typical organizational response to such a problem would be to bring together domain experts with different responsibilities and skills into a "war room," in order to try to bring clarity to how the problem emerged, identify the root cause, and arrive at an optimized strategy not just for addressing it, but ensuring it doesn't recur.
The problem, though, is that even in such a scenario, much of the information the domain experts would need remains inaccessible—locked behind the interfaces of different management solutions. For a problem that cascades across multiple systems and services, for instance, multiple management tools (addressing resources and IT assets as varied as storage, composite applications, and virtualization hypervisors) might all come into play. Each would typically involve its own data repository, and each would feature its own specific interface. The result is that despite the collaboration of domain experts, it's still relatively slow and costly to identify and weave together cross-domain information in the context of the unique problem being addressed.
Via Jazz-based solutions, this situation can be significantly improved. Because Jazz supports cross-solution data links, it's possible to create a single, unified view of relevant, in-context data—the ideal resource for troubleshooting purposes. This helps to diminish both the time-to-resolve and cost for each such problem.
How much? While cases will naturally vary, suppose that domain experts typically spend 20 hours per week in the “war room” diagnosing and troubleshooting such issues. Each domain expert represents an hourly cost of $50, with on average 10 such experts convened at one time. This means the organization spends more than half a million dollars—$520,000—every year on labor for complex problem resolution.
Should this organization move to Jazz-based solutions, they could achieve a single integrated view of multi-domain, multi-vendor data—all displayed in context for support staff. In this scenario, the numbers become much more attractive as those domain experts that are not relevant to the problem could quickly be excluded. IBM estimates drawn from its extensive case history suggest that half as many experts are needed on average, and because data is provided in context, issues are typically resolved in half the time.
As a result, the total cost of problem resolution falls to $130,000—a savings of 75%.
Scenario: Integrating solutions over a three-year period
Achieving those results, of course, presumes cross-tool integration exists. And in this area, too, the standardized, OSLC-based design of Jazz for Service Management can pay heavy dividends.
Consider the typical costs of implementing such integration by way of solution-specific APIs. IBM estimates that, on average, the project timeline for integrations is two weeks, with a projected cost of consultative services (e.g. Systems Integrator, or VARs) in the range of $30,000. Given these estimates, the cost of labor per day can then be implied to be $3,000/day.
And for many organizations, the total number of such integrations needed every year is fairly high, driven by the major releases and upgrades to vendor applications and APIs. The result is that as new application versions are deployed, new integrations will typically need to be created. For an organization that pursues only 10 new integrations per year, then, the total labor cost required is $900,000—a figure almost any organization will find uncomfortably high.
Given Jazz-based solutions, that figure can be rendered far more acceptable. In the first year, the 10 new integrations, at two weeks per integration, may still require $300,000, but each year after that, even given new version releases, a from-scratch integration typically won't be required.
Instead, a one-day validation process per integration will usually suffice, dropping the cost by 90%—all the way down to $30,000 to validate all 10. Since the same is true the following year, the total three-year labor cost under Jazz for Service Management comes to only $360,000—a drop of more than half a million dollars, to achieve superior functionality.
Scenario: Maximizing business service uptime
A third lens through which to view Jazz strengths is business service uptime. For client-facing, revenue-generating Web services, the phrase "time is money" is particularly apt. And going forward, as more and more organizations leverage software-driven services as a major or primary business model, that phrase is only going to become more and more popular.
This being the case, it follows that the faster that failed services can be restored, and customers can carry on conducting transactions as normal, the better the business outcome organizations are likely to get. This is easily demonstrated via some simple calculations.
Suppose the e-commerce impact of downtime for a particular service is $10,000/hour (a conservative figure in many cases). Also suppose that, on average, the total downtime for the service is 1.6 hours per week (an estimate courtesy of Dun and Bradstreet¹). As long as that situation persists, the total revenue lost to the organization per year comes to $832,000.
However, per the first scenario, we know the time required to troubleshoot even complex technical problems can be expected to fall considerably via a Jazz-based solution portfolio—in fact, by half. So we can also cut the annual losses due to downtime in half: $432,000.
Each of the three scenarios provides compelling return for each affected audience: IT support staff, integration team, and line of business manager. But consider also that these three scenarios are directly linked to each other, and only by analyzing the collective impact of all three do you get a true sense of the total business value Jazz for Service Management can provide.
In sum, the total three-year business impact is almost $3 million saved—the consequence of reductions in mean time to resolution, time to market for new capabilities, and labor costs, as well as recaptured revenue from improved uptime. And that $3 million, which would otherwise represent the cost of lost opportunity, can now be invested toward any business priority, such as the development of new services and capabilities, which accelerate business growth, and improve client value.
¹ Dun & Bradstreet reports that 49 percent of Fortune 500 companies experience at least 1.6 hours of downtime per week. That translates into more than 83 hours per year. (Source: Henry Martinez, “How Much Does Downtime Really Cost?” Information Management, August 6, 2009).