The key to maximizing Azure cost savings is to optimize in a logical order. For example, rightsizing VMs before purchasing Reserved VM Instances will ensure you don’t pay more than necessary for a committed use discount. It’s logical, but not every under-pressure DevOps professional always thinks logically!
Most businesses looking to migrate to the cloud typically do some online research before starting their cloud journeys, so it’s a little surprising that so much money is wasted on Azure Cloud services unnecessarily. Awareness about overspending in the Azure Cloud has been around for years; and, each time a report about cloud spending is published, it often dominates online news services.
Recent headlines about business paying for capacity they don’t use, overshooting budgets, and receiving cloud bills two to three times what was expected can’t have escaped the notice of CIOs and CFOs. A common reaction is to investigate whether their businesses are spending more than necessary, but CIOs and CFOs are likely to delegate the role to a systems administrator or DevOps professional.
Systems administrators and DevOps professionals usually have enough on their plates without having to look for Azure cost savings. However, in order to address the “request from above”, they may simply look for the most obvious overspend and deal with it. This isn’t always the best way to maximize Azure costs savings if the cost-saving exercises aren’t performed in a logical order.
Four ways to achieve Azure cost savings
To demonstrate how a logical order can maximize Azure cost savings, let’s look at four common ways to save money on Azure - terminating zombie assets, rightsizing VMs, putting non-productions VMs on a schedule, and purchasing Reserved VM Instances (you can find more in our eBook: “10 Best Practices for Reducing Spend in Azure”).
Terminating zombie assets
The term “zombie assets” relates to infrastructure components that are running in your Azure Cloud environment but not be used for any purpose. For example, they could be VMs that were once used for a particular purpose, but are no longer being utilized and have not been turned off.
It makes sense to terminate zombie assets before rightsizing VMs, putting non-production VMs on a schedule or purchasing Reserved VM Instances, because placing this exercise to achieve Azure cost savings anywhere else in the order could result in an unused VM being overlooked.
One recent report about cloud spending found 40 percent of VMs were over-provisioned by one size or more. The price of VMs doubles for each size, so reducing VMs by one size will result in savings of 50 percent, while reducing VMs by two sizes will result in Azure cost savings of 75 percent.
However, if you were to purchase Reserved VM Instances before rightsizing any VMs with excess capacity, you’ll be paying more than necessary for the reservations and may have excess capacity that cannot be applied to another VM “if no matching resources are found in the specified scope”.
Putting non-production VMs on a schedule
Putting non-production VMs on a schedule results in savings of around 65 percent if you apply an “on” schedule of 8.00 a.m. to 8.00 p.m. Monday to Friday. If you apply a more aggressive schedule based on actual usage, the savings can be greater.
Although it’s unlikely any DevOps professional would purchase a reserved VM instance for a non-production VM - no matter how under pressure they were - identifying all your active non-production VMs, right-sizing them, and then applying a schedule is the best way to maximize Azure cost savings.
Purchasing reserved VM instances
By now, it’s clear to see why purchasing Reserved VM Instances should come last in a logical order of ways to achieve Azure cost savings. If it comes any earlier in the order, there will be an element of waste and you’ll be paying more for Azure Cloud services than necessary.
It’s important to note that these exercises need to be ongoing in order to maximize savings. One-off or periodic exercises won’t achieve the objective of preventing overspend or reducing your Azure Cloud bill to the minimum possible because there will always be an element of waste between exercises.
Maintain Azure cost savings the easy way
If you’ve been delegated the task of finding Azure cost savings, speak with us about the CloudHealth cloud management platform. Our platform gives you total visibility over your cloud infrastructure for you to easily identify zombie assets, over-provisioned VMs, the most appropriate schedules for non-production VMs, and opportunities to purchase Reserved VM Instances.
Once you have optimized your infrastructure, you can maintain the Azure cost savings by taking advantage of CloudHealth’s automation capabilities. Simply configure the platform to alert you to unused or underused assets - or opportunities to adjust schedules or purchase Reserved VM Instances - and let CloudHealth monitor your infrastructure around the clock.
By helping you maximize Azure cost savings and then monitoring your infrastructure around the clock, CloudHealth not only saves the business money, but it also saves your time as well.