Cloud cost management: How to reduce and control cloud spend

With budgeting season on the horizon, the question of cloud spend takes center stage. According to Gartner, cloud spend is growing 20% year over year, but not all organizations are reaping their return on investment. 

• What is cloud cost management?
• What are some cloud cost management tools?
• How do you reduce cloud cost?

As implied by the name, cloud cost management (CCM) is when you try to manage your cloud costs efficiently through monitoring and centralization.  It involves checking your cloud costs (ie. memory, storage, traffic etc.) and optimizing them regularly to reduce spend where possible. It’s also known as cloud cost optimization.

So, why is this important? Well, if you’ve ever had the misfortune of finding out, cloud costs can quickly get out of control, especially when there isn’t a centralized management system in place! This can occur in businesses where individuals are spinning up instances across different teams. Things get missed, costs rapidly balloon, and by the time anyone notices, the business is saddled with a big bill. Ouch!

As with most things, prevention is the best cure, which is where a cloud cost management strategy comes in to save the day. On top of helping to clarify best practices for tools and services, a good CCM strategy helps you monitor where all your spend is going. This allows you to continuously reevaluate so you can maximize your resources while controlling costs.

Learn faster. Move faster. Transform now with courses and real hands-on labs in AWS, Microsoft Azure, Google Cloud, and beyond.

Cloud cost models refer to the different pricing models used across cloud platforms and their services. There are a number of different models, but they can easily be categorized into four types: pay-as-you-go (also known as on demand), subscription-based, reserved instances, and spot instances.

It’s likely you’ll use a combination of these, as each one has its own advantages and disadvantages. So let’s take a closer look at each cloud cost model! It’s admittedly not a riveting topic, but they’re worth your time — and money, literally — to learn about.

This cloud cost model is charged on a per-use basis. This means you get charged per unit of computing power, storage, networking or other resources. This can be great if you need flexibility, since you can scale up and scale down your resources easily. However, as you continue to add resources, the costs can quickly become prohibitive.

Fixed, or prepaid subscriptions, can be offered by cloud platforms for predetermined packages of tools and services, over a set period of time. This is often for larger companies, and bigger discounts can be given for committing to longer periods of time.

This model makes the most sense for companies that are combining several hardware and software tools and services. You’ll get the most benefit from this if you’re consistently reaching your usage capacity. However, it’s best to look for another option if you’re often falling short of this.

Reserved instances offer deeply discounted rates for committing to a specified compute capacity for an extended period of time (often one or three years). This can be up to 70% off the on demand price, so it’s great for when you have a good understanding of how much compute you’ll need for the foreseeable future.

You’re also only billed for the total capacity, not the amount of instances, so this means you can change up your instance sizes within the same family as needed. 

This would not be recommended for services where your capacity requirements are likely to change regularly, but it can be useful to use reserved instances for those day-to-day core system loads, and perhaps combine it with on demand for those peak loads.  

Get the Cloud Dictionary of Pain
Speaking cloud doesn’t have to be hard. We analyzed millions of responses to ID the top concepts that trip people up. Grab this cloud guide for succinct definitions of some of the most painful cloud terms.

Not entirely different from reserved instances, Azure, AWS and GCP all offer their own Savings Plans which require a one- or three-year usage commitment. They all have these available for compute services, with AWS also having ones for EC2 instances, and SageMaker. 

It’s worth noting that GCP’s version is called a Committed Use Discount (CUD), but follows similar commitment timelines and percentage discounts as the other providers. 

You’ll get similarly high discounts for spot instances, sometimes even higher (up to 90%). However, there’s a catch! These are when cloud providers sell off spare capacity, but can be stopped at any time. Because of this, they only make sense for processes that can be interrupted without issue, or if you want to do, say, a day of high-load processing. 

Cloud cost management tools help you monitor your usage and spend across your cloud services, with some of them making recommendations around permissions, security and ways to cut costs. Each major cloud provider has their own free service, as well as paid third-party options that make sense for multi-cloud setups. 

Amazon CloudWatch is the AWS cloud cost management tool that monitors your complete stack, which includes applications, infrastructure and services. It provides automatic and custom logs, and provides actionable insights that you can use to reduce costs and usage. It also has a number of automatic actions it can take, such as Auto Scaling, or rules based on operational changes and alarms.

Find out more about Amazon CloudWatch.

The Azure cloud management tool is Microsoft Cost Management and Billing. It works in a very similar way to CloudWatch, in that it is a suite of tools that help report on, and analyze your costs. You can also monitor your costs with alerts such as budget alerts, anomaly alerts and scheduled alerts. You’ll get recommendations on how to optimize your costs through tools such as Azure pricing calculator or Azure Advisor cost recommendations.

Find out more about Microsoft Cost Management and Billing.

As imaginative as the other cloud platforms, GCPs cost management tool is called…Cost Management. Along with reports, dashboards, budgets and alerts, you can also structure and organize your resources and permissions with this tool. You’ll also get intelligent recommendations for optimizing your costs and usage.

Find out more about GCP Cost Management. 

CloudHealth by VMware is a third party tool that offers similar services to the cloud provider cost management tools. The main difference is that it supports all the main cloud providers, as well as Oracle cloud and VMware (of course).  

As with others, you can gather reports with recommendations for optimizing your usage and costs, you can automate tasks, and you can strengthen security through permissions. The reason you may want to go with a third-party provider like CloudHealth, is if you’re a multicloud company - which is something that’s becoming more common these days. 

Find out more about CloudHealth.

There are a number of other third-party cloud cost management tools, such as CloudZero, Densify, Virtana Optimize and more. These might be worth checking out and comparing features if you have a multi-cloud setup. 

Automating AWS Cost Optimization
Using AWS cost-effectively can be a challenge. In this free, on-demand webinar, you'll get an overview of AWS cost-optimization tools and strategies, like data storage optimization.

There are a number of ways to reduce your cloud costs: auto scale your instances, reserving capacity for large discounts, identifying and removing unused instances, and rightsizing computing services. Most cloud cost management tools can help with many of these, and having a good cloud cost management strategy in place. 

Let’s take a look at some of these cost-busting options!

Many cloud providers offer free tiers of tools and services that you can try out for free before committing, or up to a certain amount for free per month (or other time period). This is a great way to try before you buy, or plan your usage around free tier limits if that makes sense for your workloads. 

This basically allows you to scale up resources when needed and scale back down again when they’re not. This helps reduce your spend, particularly for resources that are pay-as-you-go.

Using reserved or spot instances (discussed above) where you can, can save you up to 90% off the regular amounts. This won’t be possible for all of your services, but for long-running services with a consistent usage, it can work well. 

This is as simple as it sounds, remove any instances that are being underused, or not being used at all. Although sometimes identifying these instances might not be as simple, particularly in large organizations. That’s where cost management tools help - which are discussed below.

This refers to shutting down non-essential instances when they’re not needed. You may have some that run 24/7, but it’s worth diving in to see if this suits all of your instances. Being able to power schedule even a few instances could save you a lot of money in the long run. 

Right-sizing your resources, such as compute, just means making sure that the instances you choose make sense for your needs. Monitoring usage over time will help you understand if you’re under-utilizing instances, which you can then swap out for something more suited, saving you money. 

A great way to automate a lot of these cost reduction methods is to use a cloud cost management tool, like the ones discussed above. They provide reports that will allow you to identify resources you can optimize by right-sizing, or even removing completely. They can automatically scale your instances, and help alert you when things aren’t running smoothly. And most importantly, they’re often free to use. 

Creating a cloud cost management strategy for your organization is a great way to stop costs from spiraling out of control. It should include best practices for permission hierarchy, setting up new instances and resources, reporting and analysis schedules, and other relevant policies to help centralize and monitor your usage and spend. It’s also good to make sure your strategy scales well, as organizations rarely stay the same size forever. 

If you’re looking to learn more about CCM, we highly recommend checking out the following resources:

• Cost Optimization | A Cloud Guru 
• AWS Cloud Services and Infrastructure – Cost Optimization Deep Dive