Complex Systems engineering in a federal IT environment: Lessons learned from traditional enterprise-scale system design and change

摘要

The fragility created by hierarchical organizational constructs crosses over into the design of many large scale IT systems that are distributed across an enterprise. This means, that for these systems, end-to-end system design comes from the top down, creating a situation in which all fragility rises up to the largest scales of the system; this is a result of these systems being centrally controlled, often at the top of a hierarchy. In order for enterprise systems such as these to augment or repair themselves, they must undergo a catastrophic, enterprise-wide failure and be reengineered, once again by top-down direction [1]. This is the opposite of resilient system design and represents a situation where federal IT can be very inefficient. The current climate in the US of proactive and aggressive infrastructure consolidation via the Federal Data Center Consolidation Initiative (FDCCI) and the National Defense Authorization Act (NDAA) only serves to further incentivize system designers to construct extremely fragile systems at both the application and infrastructure layers. This fragility puts these systems at greater risk for enterprise failure. An example of a critical federal enterprise IT system's design and resulting fragility when perturbed (i.e., via consolidation and modernization) will be examined in this paper. Engineering guidelines from a complex systems perspective will be recommended to counter this resulting fragility. These guidelines will be from both an IT and government policy point of view and are generalized for applicability to systems engineering outside the scope of just IT systems.