What is Cloud Cost Optimization?
There are so many questions that a Cloud Architect must continuously ask of their cloud usage to keep downward pressure on cloud costs and avoid runaway bills:
- How do I reduce my Azure & AWS bill?
- How much of last month’s cloud bill was wasted? Was it 10%, 35% or even 70%?
- How do I easily identify the wasted cloud resources in my cloud architecture?
- How do I quickly search my entire cloud accounts for a specific rogue resource?
- How do I configure an automated rule to make sure it doesn’t happen again?
- Who’s job is it to find waste and save money in the cloud?
A Cloud Architect’s difficulty in answering these questions is an indicator of the size of the risk that they are suffering from the same, number one cloud challenge as everyone else: minimizing cloud costs.
This article looks into how to control cloud costs and how to arm Cloud Architects with the tools that complement their skillset.
Why is Cloud Cost Optimization Important?
The superpower of the cloud is its dynamic nature and seemingly infinite capacity - but as all good superheroes know, “With great power comes great responsibility.” The responsibility for cloud cost usually falls first onto the shoulders of the Cloud Architect. In addition to their technical role, cloud forces architects to stretch their skill set to understand cloud finance. The failure of a Cloud Architect to control cloud costs risks a large unexpected cloud bill that will damage a cloud project’s survival. If it gets really bad, a runaway cost may threaten the company’s survival.
There are many stories littered across cloud history of costs running away and people getting very large and very unexpected cloud bills in the finance mailbox. Even the best engineering companies like Adobe can get bitten to the tune of $80k per day.
In this article on The Information, cloud bills can be staggeringly large and growth can be powered by waste, not value.
The report states that “The median cloud spend for small-to-medium businesses is around $120K, and 10% of SMBs are spending $1.2 million or more”.
A 2019 “State of the Cloud” report by Flexera finds that cloud users are wasting 35% of their spend which means that if you’re an SMB spending $120K on cloud services, the size of your cloud waste is $42,000! That’s a full-time employee headcount.
The causes of cloud waste are many, but common and simple ones include:
- Turning on expensive-per-hour resources like large EC2 instances, not using them, not turning them off, even forgetting about them. This can cost thousands of dollars in a short space of time.
- Not understanding the “not so obvious” cloud costs like network egress fees or the cost of unused IP addresses or “not obvious snapshots”.
- People using the cloud who aren’t suitably trained.
Who is Responsible for Cloud Cost Optimization?
It starts with the Cloud Architect. You can’t do cloud cost reduction without them because they have to build cost reduction into the architecture (choosing the right size of resource) and they also know the most about the cloud.
The AWS Well-Architect Framework was written for Cloud Architects and one of the five pillars is dedicated to Cost Optimization.
This is the person who has learned most about the cloud, probably has a professional-level cloud certificate, and is responsible for designing how cloud resources are deployed to support an application.
The secret to cloud cost reduction is building cost-saving practices into the architecture. Nobody else can do that other than the cloud architect. The AWS Well-Architected Framework Cost Optimization Pillar tells them what they need to do.
Is Cloud Cost Optimization Difficult?
Not at all.
You don’t need to be certified in FinOps to be good at it.
You don’t need to be an accountant or a cloud cost specialist.
All you need to be is a Cloud Architect with the right actions and tools, with an appreciation of the AWS Well-Architected Framework as a minimum, to save your company a lot of money in the cloud.
Obviously, the bigger your cloud bill and the bigger your cloud wastage then there’s a point where it makes sense to focus an expert on this.
This begs the question... Do you need a FinOps team to help?
In fact, if you build FinOps too early you may be spending more on labour costs than you save in cloud costs.
The Cloud Architect implementing some sane cost practices is a good foundation for future FinOps if it becomes applicable.
How Often Do I Need To Optimize?
Just like life, you get back what you put in.
Best cost reduction results are achieved by maintaining constant downward pressure on cloud costs.
Every hour, every day, week and month of the year.
Humans aren’t great at this repetitive stuff and so the importance of a tool to do the heavy lifting is significant.
What Tools Can I Use to Reduce My Cloud Bill?
Yes, each cloud has native cloud cost tools which are free to use. They are essentially dashboards of charts to show you your raw costs. There are third-party tools that enhance these native tools to make it even easier for humans and non-accountants to make sense of all this data.
Hyperglance is a unique tool that combines a visual map of your cloud and overlays cost data on top of the map so it’s in context and makes sense. Without this, you have to switch from screen to screen and tool to tool to make sense of complex clouds.
Cloud Architects that use Hyperglance find that cost management is drastically simplified meaning they can’t imagine going back to just using the native cost tools.
Hyperglance & Cloud Cost Optimization
If you're looking to improve your cloud cost optimization, Hyperglance is the perfect place to start:
- Hyperglance is shipped with hundreds of built-in & customizable checks, all designed to help you enforce policy and reduce your cloud bill
- The checks run continuously, and are based on best practises and industry frameworks (CIS, NIST, NIST 800-53, NIST 800-171, AWS Well-Architected, HIPAA, PCI DSS, & FedRAMP)
- Checks can be used to trigger notifications, and automate fixes using AWS SNS or Azure Event Grid
- Hyperglance is self-hosted, deployed through the AWS & Azure Marketplaces, in Kubernetes, or installed on your own instance/VM