Cost Optimisation - methods of reducing cloud costs - REVIEW

In Review

Over the last couple of week’s I have discussed 8 methods of cloud consumption optimisation to generate cost reduction of your cloud monthly spend. Every little bit of savings helps the bottom line, but reality is, any change for the better achieves savings as soon as the task is executed in the cloud – which is based upon hourly consumption charges – so if you stop consuming, or consume less, then costs reduce straight away.

Furthermore, reducing spend in a month ensures that the current month bill will be lower, but every month afterwards will equally see consumption less – thus savings over a year become substantial.

So how much savings with each method discussed is possible –

1    Scheduling - opportunity to save up to 70% on compute and DB instances, especially dev workloads

2    Demand scaling – scheduling for prod workloads during business hours – perhaps 70% savings

3    Compute reservations - forecast consumption demand and get on average 40% discount

4    Database reservations - lock in long term database computation to achieve average 40% savings

5      Right Sizing – review consumption needs with aim to halve instance provisions – thus halve cost

6    Enterprise Discount Plans - multiyear commitment to gain multi percentage savings across all consumption spend

7    Latest versions – migrate to latest instance options, gain improved performance at a lower hourly cost

8    Cloud Native – deliver business functions via server less capabilities to potentially achieve up to 90% cost reductions on all compute consumption

In my experience and rough estimation, if your cloud has none of the above optimisation opportunities in place, then a combination of all or most of the above will see a total monthly on-demand bill reduced by 20% to 40%.