Explaining Adjustments Costs in DevOps

/, DevOps, Uncategorized/Explaining Adjustments Costs in DevOps

Explaining Adjustments Costs in DevOps

Coaching a DevOps implementation is a challenging proposition for a number of reasons. One of these reasons is not only getting the business side to buy in, but to keep them on board throughout the entire journey. The truth is that many organizations, vendors and coaches/consultants don’t fully appreciate just how steep the mountain is that must be climbed in order to bring about the culture change required to successfully implementing DevOps.

The implementation of DevOps will need to account for the ‘adjustment’ issues that almost always occur when making any organizational change. By adjustment issues we
mean the loss in performance that could be incurred before DevOps is fully implemented and functioning as intended.

But estimating the nature and extent of any implementation issues is notoriously difficult. This is particularly true because more often than not, the difficulty in large scale change (cultural) is greatly under calculated. When discussing DevOps-related change, we can attempt to use Bruce Chew of Massachusetts Institute of Technology’s ‘Murphy curve’ argument that adjustment ‘costs’ stem from unforeseen mismatches between the new technology’s capabilities and needs and the existing operation. In DevOps this would be unforeseen mismatches between DevOps cultural needs and the organizations existing culture.

Implementing DevOps changes, processes and tools rarely behave as planned and as changes are made their impact ripples throughout the organization. Below is an example of what Chew calls a ‘Murphy curve’ applied to DevOps. It shows a typical pattern of performance reduction as DevOps is introduced. It is recognized that implementation may take some time; therefore allowances are made for the length and cost of a ‘ramp-up’ period.

However, as the operation prepares for the implementation, the distraction causes performance to actually deteriorate. Even after the start of the implementation this downward trend continues and it is only weeks, indeed maybe months, later that the old performance level is reached. The area of the dip indicates the magnitude of the adjustment costs, and therefore the level of vulnerability faced by the organization.

Murphy Curve

By | 2016-12-06T10:53:20+00:00 December 6th, 2016|agile, DevOps, Uncategorized|Comments Off on Explaining Adjustments Costs in DevOps

About the Author: