Leveraging Enterprise Architecture to Drive IT Service Cost Transparency

A Case Study

information governance and enterprise architecture

Like so many other companies, VMware is on a long-term journey to transform how IT delivers services to our line of business partners. Before I lose everyone with what at first blush sounds like elevated nonspeak let me explain further. The goal of our IT organization is to evolve to the point where how we manage IT exactly matches how our end users experience IT. No one inside or outside of our company actually consumes a server, a network switch, or a storage volume directly. Rather employees consume services such as e-mail while external customers experience IT in things like order processing or working with our VMware.com website.

There is much more to our transformation story than simply organizing around services, but the focus of this article is to highlight the invaluable role that enterprise architecture solutions can play in both the initial design of services and the ongoing operationalization of those same services. The genesis of our current focus on service definition has been our IT Business Management rollout. As part of my role as the owner of VMware’s IT service management transformation, I have been leading an effort to adopt vRealize Business Management (see figure 1) as the technology solution that allows us to achieve full transparency over the cost of IT services. While not talked about nearly as much as things like automating the life cycle of services; achieving full cost transparency should be part of any program focused on transforming the way in which IT delivers services to the business.

leveraging enterprise architecture management suiteDEFINING SERVICES WITH EA TOOLS

The path we traveled in terms of defining services first involved reorganizing IT into 11 core IT functions. Over a series of several months, the IT leadership team, including our CIO, met to review how we were organized and how that organization served the end goal of delivering well-defined technical services that then contributed in some fashion to a higher level abstraction of business services. Once the 11 core IT functions were defined, the hard work of further defining the technical services that were delivered across IT started.

There are so many activities that teams perform with so many interdependencies that distilling all this activity down to a set of technical services at first seemed overwhelming. To help, we leveraged the BOST Toolkit™. You may already be familiar with the framework that is part of the toolkit, but for those who are not BOST is an acronym that stands for Business Operations Systems and Technology. The framework addresses enterprise architecture considerations of blueprinting an organization from different points of view in order to understand how a unit functions today in order for a team to better understand the opportunities to optimize for the future.

The BOST framework (see figure 2) allowed us to map the relationship between all functions and in the process to discern and define the technical services that are generated by the 11 functions. As a result of this process, 40 technical services were identified. The benefits to an IT business management initiative of having this level of transparency into how IT works are numerous. First, it provides a strong basis for building out a cost model that accurately reflects both cost pools and the relationship of these cost pools to the technical services for which costs must be attributed. The level of detail and the defined flow of activities across teams also helps in designing an architecture that allows for ongoing metering of costs and user consumption across the services that IT delivers.leveraging enterprise architecture BOST frameworkBeing able to accurately capture both costs and consumption is critical to the successful rollout and long-term sustainability of any IT business management initiative. Remarkably, as critical as it is to develop this level of transparency around IT operations in order to effectively meter costs and consumption, it is also an area where many organizations stumble badly. They do that by taking a piecemeal approach rather that starting with a “big picture” view and, as a result, plant the seeds for heartburn and heartache to their initiative downstream.

Often the failings of this step are not fully realized or appreciated until months later when the organization struggles to achieve a credible allocation of costs to services that can stand up to the scrutiny of line of business stakeholders who question the validity of the costs that are being attributed to their organization. Defining an architecture that allows for the proper collection of both costs and consumption is critical but equally hard to do. Leveraging solutions like BOST can definitely improve the odds of successfully accomplishing those objectives.

ANSWERING AN AGE-OLD QUESTION WITH EA TOOLS

In any service-oriented effort there are two early and thorny problems IT teams are forced to confront.

  1. The first question is: “Where do services come from?” Embedded in this question are other question such as “Who decides?” and “What process is used to decide?” Hopefully the last section gave you a good sense of how we at VMware tackled that problem.
  2. A second equally challenging question is: “Where do services live once defined?” This may not immediately be easily understood as a problem. But if you step back and think about it, you’ll see why this is an important question. Once defined, you’ll want every system in IT that needs a view of services to share a single common view of those services.

It’s less than ideal that once defined everything that leverages your service definitions has a different understanding of what makes up the service. For example, your ITBM solution has a different understanding of what makes up services than the systems you use to report on SLAs for those same services. If you have different understandings across systems that should have a common understanding, when you have discussions across IT or between IT and line of business partners around multiple dimensions of a services, you’ll have the classic apples and oranges problem. Inevitably, definitions that are not in sync will cause you problems that may also, ultimately, undermine the credibility of the team in the eyes of your business partners.

To solve this problem, we decided to leverage the BOST Portal on the Troux Source© repository as the ongoing source of truth for service definitions, attributes and meta data that needs to be common among multiple IT solutions. So today, our vRealize Business solution for ITBM, our ServiceNow, and our HP Project and Portfolio Management solution all leverage Troux as the single source of truth for all things services. This solution will also serve as an integration hub for service-related artifacts that are managed within one solution but need to be shared with other solutions. The BOST Portal provides the means to link these service definitions to the BOST Reference Architectures and to pick up all planned changes to those architectures and any impact on service definitions. It also connects our IT services to the consuming business functions through the linkages defined in BOST.

SUMMING UP

As mentioned, the outcome of the work done with the BOST framework was the definition of technical services. As part of the same process, the IT team also defined the business services that consumed technical services. With all of this information available, the team took the next logical step of implementing and executing a cost model with vRealize Business. The model is still in early days and will continue to evolve and be refined with an emphasis of fine-tuning both the collection of cost and consumption data. Beyond that we implemented a process leveraging Troux Source that will allow us to keep service-related artifacts in sync across the entire organization.

About Tim Pietro 1 Article
Tim Pietro is IT director, service management office & governance, at VMware.