Fresh off the holidays, many are looking to reclaim order and predictability in their lives. Yes, the year-end marathon of celebratory office lunches and staff vacations can leave one in a fog yearning to return to daily structure and certainty. Reflecting on the holiday season can give one pause to ask: “Did all of those DEV and TEST compute instances stay running while everyone was out-of-office for the holidays – Yikes!”
While the calendar advances, besides looking back, we also look ahead at financial plans with a sharp eye toward discipline and setting the pace for months ahead.
We all know the many advantages of the flexible, consumption-based, hourly pricing model for Amazon EC2 on-demand, as well as other powerful AWS services. EC2 resources can often be the dominant consumption point in many DEV and TEST environments mentioned above in our momentary lapse of remorseful contemplation.
As intuitive as it sounds, there are significant savings to be gained by simply shutting down non-production resources, while not in use. To illustrate the potential magnitude of savings, I have listed several key assumptions and built a hypothetical budget model for a small development team:
We can agree, the savings are significant if we don’t have developers and testers working around the clock, but how do you avoid any holiday-induced forgetfulness.
With a proper AWS tagging strategy, one can easily separate non-production and production resources. After resources are tagged for technical and automation, we can implement a full environment approach to fit any workflow. Whether it’s scheduled on/off times through Lambda or on-demand start-up/shut-down by pressing an IoT button – ensure your EC2 costs are optimized this new year.
Ian Gaydos is a technology leader with over 15 years of experience in application development, architecture and strategy. He’s currently a member of CSC’s AWS Global Solution Center of Excellence focused on cloud solutions.