One of the biggest complaints of enterprises that move their software to the public cloud is the associated costs of doing so, which often add up to more than they’d first anticipated. That can be a problem, since one of the main reasons for moving to the cloud in the first place is to reduce information technology costs.

With that in mind, Google said today it’s making adjustments to the way it prices its Compute Engine service in order to make things more “fair and transparent.” Compute Engine is Google’s basic infrastructure-as-a-service offering that allows clients to run workloads on its physical hardware.

“You shouldn’t need an advanced degree in finance to get the most out of your cloud investment, and you definitely shouldn’t have to worry about your cloud provider covering up costs under layers of complexity,” Google Senior Product Manager Manish Dalwadi said in a blog post.

To make things clearer, Google said it’s extending its “committed use discounts” which are offered to clients with “stable and predictable” workloads and provide discounts of up to 57% on certain kinds of virtual machine types, in return for a commitment to use them for a term of one to three years.

Committed use discounts now support graphics processing units, Cloud Tensor Processing Unit pods and local solid state drives, with discounts of up to 55%, Google said. The discounts are available in all Compute Engine regions and support a wide range of GPUs.

In an interview with SiliconANGLE, Dalwadi explained that Google was extending its committed use discounts as it has become one of its most popular pricing models. One of the main advantages is the versatility of these discounts, with customers able to reserve as few as just 10 virtual CPUs, or as many as 10,000 should they have workloads that require it.

“Many other clouds have more limitations in the types of virtual machines,” Dalwadi said. “[With Google] you don’t have to manage the utilization rate, so it really drives a lot of flexibility.”

As well as the new discounts, Google is also introducing what it calls “capacity reservations” for Compute Engine. With capacity reservations, customers can reserve compute resources in certain availability zones for specific time periods. The idea is to ensure compute capacity is available when customers need it, and should be especially useful for clients that can predict spikes in demand, Google said.

“You can create or delete a reservation at any time,” Dalwadi said. “Reservations consume resources just like normal VMs, so any existing discounts you may have apply automatically.”

Dalwadi said the expanded committed use discounts and new capacity reservation options are available in beta starting today, with general availability expected in the coming months.

Besides these options, Google also offers what it calls “sustained use discounts” for Compute Engine, which apply to customers who use the service for a sustained period of time; and “custom machine types,”, which give customers the option to define exactly how much compute and RAM they need.

With reporting from Robert Hof

Image: Mohamed_Hassan/Pixabay