The pay-per-use model of public cloud services promised dramatic savings. Companies migrated to the cloud to avoid significant capital expenditures from procuring data center capacity.
Whether it’s the lift and shift approach or building a cloud-native application, cloud costs are bound to get tricky. Even tech giants with entire departments dedicated to this area find controlling and reducing cloud costs challenging.
Knowing your costs and where they come from isn’t going to reduce them magically; it’s a good start, but you still need engineering resources to implement the changes. And not just once, but regularly – whenever you see a savings opportunity, identify a peak usage scenario, or discover an idle virtual machine left running.
Is there another way? Keep reading to find out why automated cloud cost optimization is the answer.
Why is controlling cloud costs so challenging?
Most teams struggle to control their cloud costs because they have more freedom to spin up new instances and experiment with deployments without any prior planning and financial supervision.
Some common reasons why cloud costs spiral out of control are that companies overlook the risks of the pay-per-use pricing model, have no visibility into their costs, and don’t budget for the cloud and let their bill surprise them each month.
5 cloud cost control challenges
1. Cloud costs are always changing
Predicting cloud expenses is hard because it requires calculating the resources you will use throughout the next month, quarter, or year.
During the holiday season in 2018, Pinterest had to pay AWS $20 million on top of the $170 million worth of cloud resources it had already reserved because traffic spikes increased the cost beyond what was originally estimated.
2. Resource demands never stay the same either
Using the public cloud is all about striking a balance between cost and performance. Traffic spikes can either generate a massive and unforeseen cloud bill if you leave your check open or cause your application to crash if you put rigid limits on its resources.
3. Cost visibility is harder than it seems
Managing costs in the cloud is a challenging task, especially when there are numerous cloud providers in use and various types of resources being employed.
4. Cost management is even harder in multicloud scenarios
IT teams that use multicloud combinations need to account for the costs of several different public cloud providers at once. It’s like doubling or tripling the work you’re doing for one cloud – and there are no shortcuts here.
5. Managing the cloud manually is a road to nowhere
Teams often end up locked in a cycle of analyzing their setup and allocating costs, finding better options for application placement, migrating applications to better resources, and checking whether it’s all working as expected. This has to be done on a regular basis and not just once.
This is where cloud cost optimization helps
Cloud cost optimization refers to managing and reducing a company’s cloud spending. No wonder this approach has become increasingly important in the current economic climate. The best way to understand what optimization is all about is to know what tactics it offers to teams looking to control their cloud spending.
Here are a few examples:
Selecting the right type and size of virtual machines to match application demands but avoid overprovisioning,
- Automatic scaling (autoscaling)
- Resource scheduling
- Removing idle, unused resources,
- Spot instance usage.
Not only does optimization help you achieve all of these things, but it can make the process automatic – without adding repetitive tasks for engineers. Some things just aren’t supposed to be managed manual, and cloud cost optimization is one of them.
3 examples of automated cloud optimization
1. Picking the right virtual machines for your applications
Virtual machine rightsizing and type selection can reduce a cloud bill dramatically, especially if compute is your biggest expense. But it’s not easy to pick the right instance; AWS alone has some 400 different EC2 instances that come in many sizes.
Similar instance types deliver different performance levels depending on which provider you pick. Even in the same cloud, a more expensive instance doesn’t always come with higher performance. This is where automation can help!
Instead of analyzing your application demands, picking from instance types and sizes, and exploring the cloud provider’s pricing models, you can let an automation engine make a choice for you and automatically move your application to the said resources to help you start saving on the cloud immediately.
Success story:
The Indian social media giant ShareChat uses CAST AI to replace suboptimal machines with new ones and moves applications there automatically to help them quickly reach an optimal state.” That way, our infrastructure stays optimized, thanks to the right configurations and choice of the right machine types, saving us considerable resources. We are already witnessing significant savings and expect to achieve more than a million in annual savings,” said Jenson C S, Engineering Manager at ShareChat.
2. Autoscaling to meet demand changes
If you’re running an application, you need to prepare for sudden traffic spikes and scale things down when the traffic is gone. Manual scaling of cloud capacity is difficult and time-consuming, leaving you with little time to explore cost savings.
When demand is low, you run the risk of overpaying. When demand is high, you’ll offer poor service to your customers.
That’s where autoscaling comes into play. Autoscaling solutions do it all automatically, so all you need to do is define your horizontal and vertical autoscaling policies, and the optimization tool will do the job for you.
Success story:
Using third-party autoscalers over solutions cloud providers offer was a smart move for the ad exchange platform OpenX. The advantage of using the CAST A autoscaler was that it could handle variable pricing in real time. “With CAST AI, you have very accurate and specific information that comes from the pricing API. It accounts for the region, the instance type, and so on, making decisions based on very accurate pricing,” said Ivan Gusev, Principal Cloud Architect at OpenX.
3. Managing spot instances automatically
Spot instances are virtual machines that offer up to 90% of savings off the on-demand rates, but you can lose your spot instance at any time when the provider reclaims it. You need to make sure your application is ready for those interruptions and have a plan in place when your spot instance goes away.
Automation opens the door to the sustainable and safe use of spot instances. An automation solution can provision spot instances and – if they get interrupted – quickly replace them with other spot instances. Some platforms like CAST AI offer a fallback mechanism that keeps applications running even if no spot instances are available.
Many companies hesitate to use spot instances due to the potential interruptions and miss out on incredible potential savings. The marketing automation platform Iterable was one of them – until the team tried CAST AI’s automated spot instance management. “People running Kubernetes clusters that don’t utilize spot instances would benefit a lot from CAST AI – or anyone who feels like they’re overprovisioning their infra when they don’t need to be so. It’s a really unique solution,” said Jason Sanghi, Staff Software Engineer, SRE at Iterable.
Automation is the future of cloud cost optimization
Traditional methods of cloud cost tracking and reporting can only get you halfway there. Cloud cost management platforms reduce costs, but automated cloud optimization can save even more.
Our data show that customers who use automation save 43% of their cloud spend on average – with spot instances, this figure rises to 60%.
Discover how CAST AI, the world’s leading cloud optimization platform for Kubernetes, can help your company save money. Request a demo today.
