Strengthening harmonization of Development (Dev) and Operations (Ops) silos in IT environment through systems approach

This paper presents a model based on cybernetics and systems thinking principles to overcome the challenges of strengthening the harmonization of Development and Operations functions in an enterprise. This was as a result of the challenges faced in applying industry standard tools/frameworks to complex problem. DevOps is an acronym for Development (Dev) and Operations (Ops) of information technology (IT) systems and applications. DevOps paradigm emerged as a response to the growing knowledge that there exists a gap of 4Cs (communications, co-operation, culture and collaboration) between what is usually considered IT development function and what is generally considered IT operations function of an organization. This gap often widens and manifests itself in many forms such as adverse competition, team conflict, inordinate delays and inefficiency of business applications/IT systems. There is a “Wall of Confusion” between IT development and operations as stated by Andrew Shafer and Lee Thompson. This “Wall” is caused by a combination of conflicting motivations among people, processes, and technology/tooling. Hence there is a pressing need for strengthening the harmonization of Development and Operations functions of an IT organization. IT Development (including testing and assurance services) professionals often tend to come with an outlook where change is the thing that they are mandated to accomplish. The business folks depend on them to respond to fast changing business and/or technology needs. Because of this intimate relationship, Dev are often expected to bring in as much change as possible. IT Operations folks tend to come with an outlook where change is the enemy. The business folks depend on them to keep the lights on and deliver the services that make the business money today. Operations folks tend to resist change as it undermines stability and reliability of IT systems/applications. Many times we heard that lion's share of all downtime is due to those self-inflicted injuries known as changes? To make matters worse and things more political and complex, development and operations teams generally position into different parts of a company's organizational structure (often with different General Managers and competing corporate politics) and often work at different geographic locations. According to a Gartner report[1], "Through 2010, Gartner expects that most IT organizations will continue to focus on the basic IT processes of application development, project management and service management. They will fail toextend this process focus to all activities, and will fail to change organizational values and behaviors.". There was a pressing need to dive deep in to this complex problem with a systems thinking approach. “The model leveraged here is called SNACTM. This stands for Stakeholders, Needs, Alterables and Constraints. This model is used to increase variety in the system in focus. Stakeholders are identified in categories like internal and external to the program. This involves identifying all stakeholders in the program, their needs, interests and influences. The Stakeholders, their needs, constraints and the alterables that are introduced to overcome the constraints are identified and considered. Alterables are the things that the organization believes it can alter or change.”[13][15] This model enables to identify key objectives that are common to both Dev and Ops. This also involves identifying key themes of DevOps which form the levers of transformation that is being delivered. The stakeholders, their interests and influences help focus on the key themes of DevOps strategy. This helps the program managers/directors to focus on key themes that need to be in focus to respective stakeholders at right time when they are needed the most during delivery lifecycle. The paper covers an overview, the theoretical basis on which this model has been built. It also covers the model itself in detail. The paper also illustrates a case, which used this concept, conclusion and future direction for the framework. The model was deployed successfully to assess the feasibility and craft DevOps strategy of a large IT program of customer contact transformation for a large telecom company in EMEA (Europe, Middle East and Africa) region providing telecom services and products to tens of millions of customers. The results of application of the aforementioned systems science model to address the problem are also illustrated.. The outlined model here, helped in the identification of levers to protect DevOps stakeholders interest and enhance effectiveness and efficiency.