Business-driven IT management

Claudio Bartolini, Cesare Stefanelli
2011 12th IFIP/IEEE International Symposium on Integrated Network Management (IM 2011) and Workshops  
these could be about revenue, cost, inventory turnaround time, etc. Monetized business metrics are of particular interest, since they are 15 understood in most business settings and allow algebraic operations such as addition. 2. Select ( technical) performance metrics in the context of the IT management scenario of interest. Scenarios include IT service management processes, autonomic computing platform self-management, software engineering, IT projects, and strategic planning, IT governance,
more » ... tc. 3. Model the relevant entities in the scenario of interest, their attributes and their relationships; and quantify IT-business linkage, i.e., estimate the impact that the IT scenario solution has on the business metrics, or, the other way around, how business metrics can lead to IT parameters. 4. Validate model, making required enhancements in the model itself and in its associated IT-business linkage quantification. 5. Use the validated model to support decisions concerning IT solution in scenario of step 2. 6. For the scenario of interest, evaluate gains in business results. Compare gains to business goals. In case discrepancies are unacceptable, make adjustments in the IT solution. Note that the IT solution could still be in the making, as in a proposed project. In this case, the above steps would evaluate possible business gains (e.g., return on investment, ROI) when the project is realized (thus, step 6 may apply to its design). Notice also that by automating the above steps and by looping through them, one does in effect get a BDIM control loop. When all steps are automated, BDIM control may be encapsulated into autonomic computing infrastructures to enact online, on-the-fly self-management. The focus of this dissertation is however on BDIM solutions 16 aiming at providing decision support to human agents rather than on autonomic solutions aiming at taking the human out of the loop. This way of operating is particularly appropriate when tackling decision problems that IT managers and IT staff face in IT service management processes, which is our application domain of choice for this thesis. Step 3 in the methodology described above is where the hardest challenges often reside when building and operating BDIM solutions. To that end, an appropriate BDIM IT-business linkage modelor simply, a BDIM modelmust be used. Models for solving a BDIM problem essentially describe relations between business and IT measures. For instance, one such relation could be a function that yields revenue from IT service availability estimates. BDIM model features and resources are to be determined by research efforts which include constructing and validating BDIM models while simultaneously studying IT governance and IT management best practices and methods for eliciting knowledge and policies related to IT and business decisions. The resulting models also depend on the situation and goals of the model"s intended usagesuch as in the case of autonomic computing or in a decision support tool for IT managers to use. 17 2 State of the art and research challenges in Business-driven IT management The purpose-driven definition of BDIM that we have adopted makes it so that many works presented in the areas of distributed systems, network and system management, economics of IT and organizational behavior can be considered as BDIM applications ante litteram. By studying those contributions, along with other more recent ones whose authors consciously position in the BDIM research agenda space, we realize that many interesting challenges remain open, in the two orthogonal dimensions of modeling for BDIM, and advancing the state of the art at the intersection of BDIM with the disciplines of autonomic computing, IT service management, and IT governance. We present an overview of these challenges, and decide to focus our contribution in BDIM decision support in IT service management, and contribute decision theoretical framework and models that bring the concepts of business impact and risk to the fore, as we will see in later chapters.
doi:10.1109/inm.2011.5990530 dblp:conf/im/BartoliniS11 fatcat:eiwspuykqzf6nhdfvwozrxecby