Cloud compute expenses consume a healthy portion of a technology or SaaS company’s annual budget, generally from 4% to 12% yearly. Best-in-class companies that are cloud-native and all-in can come in at around 1% of revenue, but they are the exception, not the rule. According to AWS, for highly technical projects where the application is the business, the expense can reach 40% of revenue.
Many resources are consumed for Cloud hosting, which can be better deployed elsewhere.
- Lengthen runway
- Increased free cash flow
- Increased agility
- Lower OpEx and increasing related metrics
How much can I save?
Steady state The average stable growth company can save 10% – 15% of their annual AWS Cloud expense with a few simple changes that have no negative impact on operations.
High growth Whether new product lines or start-up operations, these companies require more technical effort to optimize their Cloud expenses. It is difficult to make concrete plans when you are staring at a hockey stick curve and potentially exponential growth.
Migrating to the Cloud from on premise, AWS estimates an immediate 20% – 40% cost savings the first year of operations.
Using AWS serverless (Lambda, Fargate), where all you are managing is your code or a Docker container, you can realize up to 90% savings.
Contract for one or three years with no prepayment, partial and full prepayment, and save up to 72% off your monthly AWS bill.
Over-purchasing on the Savings Plan can leave the user with unused capacity; under-purchasing can lead to provisioning expensive spot instances. Rapid growth can muddy the waters further as long-term requirements remain uncertain.
Customers on AWS enterprise plan enjoy considerable (unpublished) pricing and, if a significant enough use case, negotiated rates and SLA.
Right size before migration
If you are doing a lift and shift or greenfield Cloud deployment, a critical step to cost optimization, and security, is to trim your network resources before migration. In other words, you are just bringing the legacy problems with you to your new deployment. Key consulting personnel that can assist here are Cloud Solutions Architect, AWS Engineer, CyberSec, and a Product Manager to oversee your migration. Network migrations are typically mission-critical to your company’s future once they begin, so it is critical to plan carefully and have top-level buy-in and frequent interaction.
Right-sizing is using the lowest cost instance while meeting your current demand. Typically you can push these servers to an 80% load before you have to scale out or up.
It is better to be a little ruthless with decommissioning than to leave things running “just in case they are needed at some point in the future.”
Instance scheduler allows you to auto-provision Cloud resources for peak hours, holidays, and marketing campaigns and then scale them back to idle when not required. Scheduling is similar to auto-scaling, with the difference being that scheduling is planned, and auto-scaling happens automatically. Scheduling can also be regional, which includes countries.
Delete Idle Resources
When there are three guys in an apartment banging out code day in and day out, it’s simple to keep on track of your AWS bill, as it is likely the difference between making rent this month or not.
As your company scales and your team of three turns into a group of 45, you can no longer track what everyone is doing when both experiments and R&D begin. No engineer or scientist ever would knowingly kill their colleagues’ test run; people don’t touch things they do not understand. These orphaned resources are fired up, forgotten about, and left to run, leading to an ever-increasing AWS bill with absolutely no reason.
A step further is zombie applications, which are abandoned deployments that are not managed, patched, or even known about. Perhaps a former employee’s pet project. Somehow these applications made it to your production or private network, and worse; they have IAM security credentials. The implications here are massive if you have a breach, and the financial fallout can be the difference between having a company tomorrow or not!
Scales your Cloud resources up and down to meet demand in near real-time. Generally, a best practice is to utilize many smaller size servers rather than larger instances. An exception to this smaller is a better mindset, clustered computing such as Kubernetes, AI, or GPU, which may have multiple reasons for running and orchestrating larger server instances.
A Cloud Solution Architect can help determine if your existing applications are suitable for auto scaling.
Between scheduling and auto scaling, you should be able to reduce your nonproduction expenses by 60% or more.
Maintaining cost visibility…for free!
- Slack; meet your team leads once a month and discuss Cloud spend.
- Monthly AWS invoice
- AWS Billing Console
- AWS Cost Explorer
- AWS Budgets function
- Total Cost of Ownership Calculator
Optimizing cost is not a one-time fix. Due to the complexity of AWS Offerings, creating a one-time solution is next to impossible. This should be treated as an ongoing pursuit.
- Cloud financial management.
- Awareness of usage and expenditure.
- Optimize over time.
- Analyze and attribute expenditures.
Want to learn how you can save money on your monthly cloud bill?
On your own
- Start with your monthly bill and your tech leads. Can you track down what everything is? Does it look right?
- Do you know what your network looks like? A fantastic and inexpensive tool is Lucid Chart, which can output your entire network design in a few clicks.
- AWS Cost Explorer is a free tool and allows you to see high-level data and trends quickly or allows you to do a deep dive into any of your data.
For a complete Cloud economic review based upon AWS Well-Architected philosophy with cost optimization guard rails in place, please reach out to CloudView Partners. We have a wealth of experience and can assist you in migrating to the Cloud and living well in the Cloud.
If you wish to learn more about cost optimization and best practices, you may find these links helpful.
- CloudCOE by CloudView Partners
- AWS, Cost Optimization Pillar
- AWS Well-Architected Labs, Cost Optimization
- AWS Cost Management Blogs
Optimize and Enhance the Maturity of Your Cloud Services, Transforming Your Program into a Cloud Center of Excellence.
Audience: CIO, CTO, VP/Director of Cloud, Technical Founder, and Organizations looking to operationalize their Cloud.
Benefits: Enable your workforce to become Cloud-savvy, or increase the maturity of your Cloud environment.
CloudView Partners exists to help organizations painlessly navigate their unique journey towards enterprise Cloud adoption. CloudView Partners believes that Cloud is the new normal, and our goal is to help companies transform by creating a safe, reliable, and low-risk path to the Cloud. Our team of highly-skilled Cloud Architects brings with them experience and lessons learned through years of practitioner expertise, allowing our clients to adopt the Cloud the right way, the first time around.