Software Engineering
5 Mins
June 28, 2024

How to Do More for Less with Cloud Infrastructure Optimisation

How to Do More for Less with Cloud Infrastructure Optimisation

Digital transformation initiatives often include the migration of in-house services to cloud infrastructure. This approach helps enterprises scale easily, enable innovation, and unlock significant cost savings. 

However, cloud migration isn’t straightforward. Sometimes you’ll have to deal with a disconnect between potential and reality. Sometimes, because of cloud complexity, you might not even achieve your target savings. If this sounds like your company, you’re not alone.

According to Gartner, global IT spending is expected to add up to $4 trillion in 2021. However, according to the Flexera 2020 State of the Cloud Report, 23% of organisations exceeded their cloud budget last year, while another 47% expected it to increase in 2021. A further 30% of respondents also estimated that 30% of their cloud spend was wasted.

This suggests that it’s essentially a balancing act that helps us access significant savings. If you’re like most companies, for example, you’ll have to find a balance investing in cost-saving and growth.

Financial control and innovation don’t have to be independent of each other. Instead, companies that invest in cloud infrastructure will also be well-placed to be digital leaders and innovators. The best approach here is to find a way for both innovation and budget control to co-exist through discipline.

In this scenario, it’s critical to understand that cost saving isn’t a standalone initiative. Instead, it goes beyond engineering and demands that the leadership team make it a business imperative.

So, how can enterprises do more for less?

How do organisations enable cloud cost optimisation without hindering innovation? 

Although boosting operational efficiency might be the focus of your migration project, it shouldn’t be your only focal point. Your team (including leadership) should determine the surface area for your technology stack, product, and services using cloud cost optimization opportunities.

This article will explore the steps taken by us at Evolve to help companies access significant savings while accelerating growth.

Step 1: Audit Your Product Suite

The first step in the process is to audit your product inventory. It’s important because you don’t want to waste time migrating legacy products with limited scope. You also want to identify and eliminate outdated non-strategic use cases. 

However, this process isn’t always straightforward. Take the time to determine which ones must stay and which ones you can do without. In the end, if you do decide to include older products, create a seamless transition plan to upgrade or update them in the near future.

Step 2. Feature Adoption and Scope

During your audit, did you come across any product features that didn’t achieve product-market potential? Are they still actively supported? If you answered yes, you have to do something about it.

Take it as an excellent opportunity to ascertain if you should keep these features or decommission them. This approach will help you gain some efficiencies over the course of this exercise.

Step 3: Homogenise Your Technology Stacks

How are your development teams using technology? What about different departments and across various roles? What are their deliverables? What type of work are they doing with the tools available to them?

The answers to these questions will help you find the best way forward. Do your homework and see if there’s an opportunity to standardise products and reduce your overall technology stack. 

However, keep in mind that the reduction of services may lead to high costs in the beginning. But it also creates an opportunity to re-evaluate your scope and request usage discounts (from your cloud services provider).

Step 4: Leverage Data Optimisation Protocols

Over the last decade, data was hyped up to be the new digital gold, and everyone wanted a piece of the action. But there was hardly any discussion about reducing data storage costs, data processing costs, and data optimisation costs. 

Going forward, we must approach data based on optimising costs. This means that you should only store data that enables full product functionality and not a byte more. 

Step 5: Strive to Establish Good Data Design

With an increase in privacy regulations, businesses are now more focused on data governance and best practices. When you’re in the cloud, be sure you have a clear need to collect and process data. You’ll also have to provide consumers with notice and a choice whenever their personal data is collected.

When you have a good data design, it’s easier to deliver robust products that align with regulatory compliance protocols and overall security. As good data design also includes minimisation, you’ll be well-placed to adhere to local and national privacy laws. 

Whenever the processing is complete, delete any intermediary data by including lifecycle management and best practices. In the same vein, establish a culture of continued data optimisation across the organisation. This approach will ensure cloud cost optimisation as your business scales.

Step 6: Manage Cloud Waste at Scale

Cloud resources are essentially like utilities, as you only pay for what you use. This means that it’s time that businesses started treating just like other utilities like electricity. For example, if you turn off unnecessary lights and heating at the end of the working day to reduce expenses, you should do the same on the cloud.

However, it’s not as easy as switching off a light. To save on cloud resources, you should focus on the following:

  • Recognise unattached resources and delete them. This approach can add up to thousands of dollars in savings depending on size, volumes, and the total amount of unattached volume.

  • Identify idle workloads and suspend them after business hours. This usually occurs in non-production environments where almost nothing happens between 6 PM and 6 AM on weeknights. The same is true all day long on weekends. This approach has the potential to yield significant cost savings.

Unattached storage is a universal problem. It’s pretty common for users to forget to delete associated volumes when they destroy instances. This is bad news for enterprises as they will pay whether they use it or not.

You can automate this process by using scripts to suspend and resume virtual machines on a schedule. However, as the environment evolves, it can be a challenge to maintain and update these scripts. So, find a user-friendly self-service solution to save time and resources.

While operation efficiency and cost-saving are the primary focus, expanding the project scope and sourcing ideas across departments can have an impact on overall cost savings. This makes it vital to make this project as inclusive as possible.

When you take on a cloud cost optimisation project, it also helps to focus your communications on a strategic initiative instead of just a cost-reduction effort. In this scenario, it’s crucial to highlight what worked and what didn’t to get everyone to buy into it.