7 use cases for Google Cloud committed use discounts

Options for committed use discounts

Users can choose between resource-based and flexible committed use discounts. Both options require you to know the trends of your past cloud usage and spending in order to realize the full savings from these options.

Resource-based

Resource-based committed use contracts are for predictable and steady-state usage. Also, an enterprise commits to use a certain amount of Google Compute Engine resources in a specific region. These commitments include the following:

• Hardware commitments for resources, including virtual CPUs, memory, GPUs, local solid-state drives and sole tenant nodes.
• Software license commitments for OS licenses.

Commitments for hardware are separate from licenses. You can purchase both for a VM instance, but you cannot purchase a single commitment for both hardware and software.

Flexible use

There's a more flexible option that enabels your organization to commit to a certain usage level for one or three years but with the flexibility to allocate these resources across different services and regions. Your organization receives a discount based on your usage cost based on the total amount of resources committed, even if you use those resources differently than you may first anticipate. It is recommended for enterprises that have predictable spend needs.

The discounts are as follows:

• 28% discount over your committed hourly spend for a 1-year commitment.
• 46% discount over your committed hourly spend for a 3-year commitment.

Keep in mind that both types of discounts only work for resources deployed using Compute Engine SKUs.

Use cases for Google committed use discounts

Explore Google committed use discounts as a standard practice rather than just for a project. Here are some use cases for the discounts.

Steady-state workloads

If a cloud project with steady-state workloads requires a predictable number of resources, that project is a candidate for a CUD. For example, say your organization is building an internal application for internal reporting with a database back end. The application requires a certain amount of computing power and storage by design. Your organization also has complete control over the user community. If this is true, it's a potential candidate for a CUD.

Predictable workloads

If your organization uses predictable workloads for long-running instances of persistent disks, it's a definite use case for CUDs because their predictability leads to cost savings.

Account for workload trends and metrics using your cloud management platform to determine which workloads qualify for savings. This information comes with being a more mature cloud operation. It could take three months, six months or even a year's worth of data on your cloud trends to show that you have predictable workloads that qualify for this discount.

Predictable growth

A cloud migration roadmap can become a powerful tool for predictable growth because it outlines your timeline for migrating workloads to the cloud. DevOps practices and automation are other tools for predictable growth because these practices provide consistent and reliable deployments.

If these practices are in place, budget and resource planning are now possible. This puts you in a better position to commit to a certain amount of usage over time and improve efficiency.

Combination with other pricing models

It is possible to use CUDs with other pricing models, such as sustained use discounts and preemptible VMs, to optimize your costs further. Making this work requires cloud teams to coordinate closely with FinOps teams. Check in with your Google Cloud contact to ensure that the latest information on discounts and pricing models is up to date.

Budget planning

CUDs can help with cloud spending and an organization's budget. Budget planning must be more prominent in your project planning, design and cloud architecture planning. Bringing CUDs into your budgeting takes a few cycles, especially if you're still building cloud cost optimization expertise in-house.

Reserved capacity

You can reserve capacity in advance with CUDs. Developers will have the resources they need when they need them. It takes a team of more accomplished cloud architects and strategists -- with some budgeting mistakes behind them -- who've put in the tools and processes to be more predictable in their cloud solutions architecture and project planning.

Multiyear commitments

Google offers discounts for multiyear commitments, which can benefit organizations with long-term plans or projects. Making a multiyear commitment requires you having data on hand of previous usage trends.