Google Cloud Platform

Committed use discounts for Google Compute Engine now generally available

The cloud’s original promise was higher agility, lower risk and simpler pricing. Over the last four years, we've remained focused on delivering that promise. We introduced usage-based billing, which allows you to pay for exactly what you use. Sustained use discounts automatically lower the price of your instances when you use them for a significant portion of the month. And most recently, we introduced committed use discounts, which reward your steady-state, predictable usage in a way that’s easy-to-use and can accommodate a variety of applications.

Today, committed use discounts are now generally available. Committed use discounts are ideal for predictable, steady-state use of Google Compute Engine instances. They require no upfront payments and allow you to purchase a specific number of vCPUs and a total amount of memory for up to 57% off normal prices. At the same time, you have total control over the instance types, families and zones to which you apply your committed use discounts.

Simple and flexible 


We built committed use discounts so you actually attain the savings you expect — regardless of how you configure your resources, or where you run them within a region. For example, say you run several instances for one month with aggregate vCPU and memory consumption of 10 vCPUs and 48.5 GB of RAM. Then, the next month your compute needs evolve and you change the shapes and locations of your instances (e.g., zones, machine types, operating systems), but your aggregate resource consumption stays the same. With committed use discounts, you receive the same discount both months even though your entire footprint is different!

higer%2Brez%2Bpic7joj.PNG

Committed use discounts automatically apply to aggregate compute usage with no manual intervention, giving you low, predictable costs. During the beta, customers achieved over 98% utilization rates of their commitments, with little or no effort on their part.

Quizlet is one of the largest online learning communities with over 20 million monthly learners.

Our fleet is constantly changing with the evolving needs of our students and teachers. Even as we rapidly change instance types and Compute Engine zones, committed use discounts automatically apply to our aggregate usage, making it simple and straightforward to optimize our costs. The results speak for themselves: 60% of our total usage is now covered by committed use discounts, saving us thousands of dollars every month. Google really got the model right. — Peter Bakkum, Platform Lead, Quizlet

No hidden costs 

With committed use discounts, you don’t need to make upfront payments to see deep price cuts. Prepaying is a major source of hidden costs, as it is effectively an interest-free loan to the company you're prepaying. Imagine you get a 60% discount on $300,000 of compute usage. At a reasonable 7% per year cost of capital, an all-upfront prepay reduces your realized savings from 60% to 56%.

Screen%2BShot%2B2017-09-21%2Bat%2B63aa2.PNG
We see great financial benefits by using committed use discounts for predictable workloads. With committed use discounts, there are no upfront costs, unlike other platforms we have used. It’s also possible to change machine types as committed use discounts work on both vCPU and memory. We have been very happy with committed use discounts. — Gizem Terzi Türkoğlu, Project Coordinator, MetGlobal

Getting the best price and performance in the cloud shouldn’t require a PhD in Finance. We remain committed to that principle and continue to innovate to keep pricing simple for all your use cases. In coming months, we'll increase the flexibility of committed use discounts by allowing them to apply across multiple projects. And rest assured, we'll do so in a way that’s easy-to-use.

For more details on committed use discounts, check out our documentation. For pricing information, take a look at our pricing page or try out our pricing calculator. To get started and try Google Cloud Platform for free, click here.