Calero Blog

4 Cost Optimization Best Practices Amazon Web Services

Written by Calero | Nov 1, 2019

While more companies move towards a cloud-first IT strategy based on the technology and cost benefits cloud services provide, that migration is not without its challenges. In particular, we are increasingly seeing two related problems: first, we’re seeing those cloud costs grow (usually unexpectedly) and second, we see those expenses being managed poorly.

Solving the latter fixes the former, and that is how Network Control can bring substantial and measurable value. By applying AWS best practices for cost optimization we are able to help a business take full advantage of the cloud while ensuring they do not over-spend or lose control over that spend. Cloudwatch, AWS’s monitoring tool is deployed and the most efficient parameters selected—and then management is notified if any action needs to be taken.

In addition to Cloudwatch monitoring, we also provide other optimizations.  The most common are:

1. Rightsizing EC2 Instances

The purpose of rightsizing is to match instance sizes to their workloads. Unfortunately, it doesn’t quite work like that because of the way in which instances double in capacity for each increase in size. If you double the capacity when you go up one size, then you also half the capacity when you go down one size. Therefore, rightsizing is only worthwhile if there are instances whose peak utilization does not exceed 45%. We work with your IT team to analyze utilization metrics to find opportunities to move workloads to different areas that better meet your needs.

2. Upgrade Instances to the Latest Generation

AWS frequently announces about how products have been upgraded or features introduced to support specific cloud services. The best way to take advantage of AWS cost optimization is to make sure you are always working on the latest—and most cost-effective--instances.

3. Scheduling On/Off Times

Scheduling on/off times for non-production instances is vital to departments that develop, test and QA. This can save customers as much as 65% of the overall non-production costs by applying an “on” schedule of 8.00 a.m. to 8.00 p.m. Monday to Friday. Keep in mind, companies can save even more if their development teams work irregular hours. In that case, purchasing smaller blocks that meet their work timing can be much more cost-effective.

4. Purchasing Reserved Instances

By purchasing these, AWS costs can very easily be reduced, however, you need to match your expected needs to the term. AWS pricing changes frequently (and almost 100% of the time costs go down) so, for example, locking in a 12-month contract with one go to market strategy in mind and then changing plans up can cause you to incur extra costs needlessly, negating any savings from that reserved instance.

We work with your team to manage any reserved instances as part of our audit/best practices analysis and will set up a proactive management schedule which weighs all the variables before making a purchase and then monitoring utilization throughout the reservation’s lifecycle.