Google Cloud discounts can reduce Compute Engine costs, but the right option depends on how predictable, flexible, or interruptible your workloads are. This guide explains the main Google Cloud discounts for Compute Engine, how SUDs, CUDs, and Spot VMs differ, and what to evaluate before you commit.
Used well, these pricing options can help you lower spend without changing the workloads you run. Read on to compare the options, understand the tradeoffs, and find practical ways to improve savings over time.
Key takeaways
- SUDs fit eligible VM usage that runs for a meaningful part of the month and don’t require an upfront commitment.
- CUDs usually offer deeper discounts, but they work best when your resource needs are stable enough to support a one-year or three-year term.
- Spot VMs provide the largest potential savings, but they are best for workloads that can tolerate interruption.
- The main optimization challenge is balancing higher savings against the risk of unused commitments or interrupted capacity.
What are Google Cloud discounts?
For this page’s scope, Google Cloud discounts are pricing options that lower eligible Compute Engine costs based on sustained usage, a term commitment, or the use of spare capacity through Spot VMs. The main options covered here are sustained use discounts (SUDs), committed use discounts (CUDs), and Spot VMs.
Compare Google Cloud discounts at a glance
| Option | Commitment | Eligible usage | Savings cited on this page | Best fit |
| Sustained use discounts (SUDs) | No upfront commitment; applied automatically based on eligible monthly usage | Compute Engine resources, including GKE clusters using Compute Engine as the underlying resources | Up to 30% | Steady VM usage that runs for much of the billing month |
| Committed use discounts (CUDs) | One-year or three-year commitment | Compute resources such as vCPUs, memory, SSDs, and GPUs, with broader coverage depending on CUD type | Up to 70% for memory-optimized machine types and up to 55% for other machine types for resource-based CUDs | Predictable, long-term workloads |
| Spot VMs | No term commitment | Unused Google Cloud capacity | Up to 91% off | Fault-tolerant, interruptible workloads |
Google Cloud Compute Engine discounts and Spot VM pricing
Google Cloud offers a range of Compute Engine pricing options tailored to different usage patterns, helping you optimize costs while taking advantage of the cloud’s capabilities. Whether your workloads are steady, predictable, or interruptible, there is likely an option that aligns with your needs. Here are the main Compute Engine discounts and pricing options you should consider:
Sustained use discounts (SUDs)
Sustained use discounts are automatically applied when you use Google Compute Engine VMs for more than 25% of the billing month. The more you use your resources, such as vCPUs and memory, the larger the discount. You can get savings of up to 30% with no upfront commitment required.
These discounts don’t apply to Spot VMs, but they do offer substantial savings for standard and memory-optimized machine types.
Pros:
- Automatic application: No upfront commitment or manual sign-up required.
- Incremental savings: The more you use, the more you save, incentivizing continuous operation.
Cons:
- Limited to Compute Engine: Applicable to Compute Engine resources, including GKE clusters using Compute Engine as the underlying resources.
- Less predictable savings: Since discounts scale with usage, predicting monthly costs can be challenging.
Committed use discounts (CUDs): resource-based vs. spend-based
Google Cloud offers two CUD models, and the right choice depends on whether you want to commit to resources or to spend.
Committed use discounts are a billing feature that rewards you with lower prices when you commit to using certain resources for one or three years. CUDs provide discounted prices on compute resources, including vCPUs, memory, SSDs, and GPUs, making it a cost-effective solution for predictable, long-term workloads.
Pros:
- Significant cost reduction: The discounts can be extensive—up to 70% for memory-optimized machine types and up to 55% for other machine types for resource-based CUDs.
- Flexible coverage: With spend-based CUDs (also called Flexible CUDs, or FCUDs), discounts apply across a wider range of services and configurations.
Cons:
- Long-term commitment required: Users commit to certain usage levels for one or three years, which may not be ideal for fluctuating needs.
- Penalty for underutilization: If you don’t optimally utilize your committed resources, your cost savings decrease.
- No stacking: SUDs, CUDs, and Spot VMs do not stack with each other in any scenario. Usage can only be covered by one type of discount at any time.
There are two types of CUDs: resource-based and spend-based:
Resource-based CUDs
Resource-based CUDs require you to commit to a minimum level of resource consumption (vCPUs, memory, GPUs, local SSDs) in a specific region for a selected machine series. It’s best suited for scenarios with predictable resource demands.
Pros:
- Deep discounts: Up to a 41% discount for one-year commitments
- Applicable to multiple services: Can cover Compute Engine, GKE standard clusters, Dataproc, and Composer 1 services.
Cons:
- Limited flexibility: You commit to a minimum level of resource consumption based on your region for your chosen term length, limiting flexibility to the defined resources and machine types. As a result, you will pay for the commitment even if it goes unutilized.
- Project-specific: Discounts are tied to a single project by default, though they can extend to all projects under a single billing account by enabling discount sharing.
Spend-based CUDs
With spend-based CUDs covering Compute Engine, you commit to a minimum hourly spend on Google Cloud resources instead of a specific resource commitment. Thus, these discounts cover a wider range of services than resource-based CUDs.
Pros:
- Wide applicability: Covers various services including Cloud Run, Cloud SQL, Cloud BigTable, AlloyDB, and more. (Full list is available here.)
- Flexibility in resource usage: Spend-based CUDs offer significant flexibility by allowing you to commit to a minimum hourly spend on Google Cloud resources rather than specific resource consumption.
Cons:
- Varied discount rates: Discounts vary by term length as well as service, which may complicate financial planning. For example, Cloud Run offers a lower CUD discount (17% for one year) than the equivalent Cloud SQL CUD (25% for one year).
- Commitment to spend: Spend-based CUDs are purchased at the Cloud Billing account level and cannot be canceled or changed.
Spot VMs
Spot VMs offer a flexible and cost-effective solution for running fault-tolerant workloads on Google Cloud. These discounted instances utilize unused capacity and can be terminated by Google at any time, but they provide the highest potential savings. Spot VMs are significantly cheaper than standard On-Demand VMs because they rely on Google Cloud’s surplus capacity.
Pros:
- Cost-effective: Significant reductions in compute service costs (up to 91% off).
- Ideal for flexible workloads: Best suited for fault-tolerant applications that can handle interruptions.
Cons:
- Risk of termination: VMs can terminate at any time, which may not be suitable for critical workloads.
- Requires management: Additional architecture considerations are necessary to handle possible interruptions.
Factors to consider when choosing Google Cloud discounts
Before choosing a discount option, you need to weigh the savings opportunity against commitment risk, workload behavior, and future growth. These are the main factors to review before you decide.
Usage patterns
Your project’s usage patterns are key in determining the most cost-effective discount. For example, if your workloads run continuously on an N1 machine, you might save up to 30% with sustained use discounts, as they are designed for VM instances with steady-state usage. However, if you purchased a three-year resource-based CUD at 100% coverage, you could save up to 55% off eligible usage.
Conversely, sporadic or unpredictable workloads may not benefit as much from committed-use discounts as they may go unutilized if usage decreases. Therefore, you need to analyze your billing cycle data to understand your usage trends.
Budget and cost management
Cost management is a balancing act, particularly in a cloud environment. Consider applying the principles of Cloud FinOps to stay on top of your budget and maximize your cost efficiency.
Before choosing a discount type, determine whether it aligns with your budgetary constraints and commitment level.
Workload types
Each discount best suits a different workload pattern. Compute-intensive workloads might rely more on vCPUs, while data-intensive tasks may demand additional memory or storage resources.
Assess your workloads and their usage patterns against the qualifying factors for different discount programs to ensure they meet your needs. For example, if you have a lot of unpredictable usage in projects, sustained use discounts may offer more value.
Scalability and growth
Finally, consider your long-term plans for scalability and growth within Google Cloud. If you expect your project to grow, look at how each discount option adapts to increasing demands on Google Cloud services and resources. Will the discount scale alongside your needs, or will it hinder growth due to a rigid commitment?
Project out future resource requirements to avoid locking yourself into a discount that does not scale efficiently with your expanding cloud presence. Over-commitment for CUDs can be just as problematic as under-commitment!
Tips for optimizing the use of Google Cloud discounts
If you want to optimize Google Cloud costs, you need to leverage Google Cloud discounts strategically. Here are a few practical ways to make the most of your cloud investment:
Assess and forecast usage accurately
Begin by assessing your current usage and estimating future requirements with as much detail as possible. Accurate forecasts help you align your commitments with actual needs, preventing over-commitment and wasteful spending.
Use tools within the Google Cloud Console (such as Google Cloud Monitoring, Google Cloud Logging, or Google Cloud CUD analyzer) to track usage patterns and plan for capacity with predictive analytics.
Leverage committed use discounts for predictable workloads
For consistent and predictable workloads, consider committed use discounts. These usually offer better savings than SUDs because you are committed to a specific amount of resources for a one-year or three-year term.
To maximize these discounts, it’s important to analyze historical cloud usage data. This will give you a better idea of stable, predictable workloads that require a certain level of resources on a consistent basis.
For more tips on leveraging CUDs effectively, consider reading: How to Leverage Committed Use Discounts (CUDs) for Google Cloud Cost Efficiency
Use sustained use discounts for flexibility
Sustained use discounts allow businesses to take advantage of discounts for real-time usage during a billing period. It’s ideal for fluctuating workloads that might not suit the rigidity of CUDs. For example, if you’re running a VM for most of the month, you can qualify for an SUD without worrying about a long-term commitment.
Optimize costs with Spot VMs
Incorporate Spot VMs for non-critical, interruptible workloads. Spot VMs can lead to significant savings since they run on spare Google Cloud capacity at a reduced cost. They can work well for batch jobs, data processing tasks, or any workloads tolerant of unexpected termination.
Leverage discount management tools
Google Cloud offers a variety of built-in tools to help you manage your discounts effectively:
- Billing Reports: This tool allows you to analyze your active discounts in one place. You can also filter visibility for your discounts as needed, allowing you to track how they are performing. This real-time visibility helps you make informed decisions about your discount utilization.
- CUD Analyzer: The committed use discount (CUD) Analyzer helps you track your CUD coverage and utilization. By providing detailed insights into how effectively you are using your committed discounts, this tool helps you identify opportunities to optimize your savings. It ensures you are fully utilizing your CUDs and not leaving potential savings on the table.
- FinOps Hub: The FinOps hub recommends cost-saving opportunities by focusing on three key optimization practices: turning off idle resources to eliminate unnecessary costs, right-sizing instances to match workload requirements for optimal performance and cost efficiency, and purchasing committed use discounts (CUDs) to lock in lower rates and achieve predictable billing.
Tip: For even greater hands-free savings, you can leverage third-party discount management tools like ProsperOps, which autonomously blends discount instruments to maximize your cloud cost savings.
Optimize your Google Cloud discount instruments with ProsperOps
When you are leveraging committed use discounts or sustained use discounts, it’s crucial to align your discounts with your usage.
ProsperOps helps you optimize these discount instruments by autonomously and dynamically adjusting commitments to real-time usage, ensuring you receive the best possible savings.
With ProsperOps, there is no friction within your operations or forced trade-offs. The algorithm runs silently in the background 24/7, without manual oversight or intervention from your busy teams.
You also get access to valuable reports with granular insights into your cloud spend. The intuitive ProsperOps dashboard with critical FinOps insights helps you track monthly and lifetime Effective Savings Rate, coverage, utilization, and more.
Book a ProsperOps demo today to learn how our autonomous discount management can help optimize your Google Cloud spend.
FAQs
What are Google Cloud discounts?
Google Cloud discounts, in the context of Compute Engine pricing, refers to three main options that lower eligible costs based on usage or commitment: sustained use discounts (SUDs), committed use discounts (CUDs), and Spot VMs. That includes sustained use discounts, committed use discounts, and Spot VMs. Each option trades off savings, flexibility, and risk differently, so the best fit depends on your workload pattern.
What is the difference between SUDs and CUDs?
Sustained use discounts are automatically applied when you use Google Compute Engine VMs for more than 25% of the billing month. You can get savings of up to 30% with no upfront commitment required. CUDs, by contrast, require a one-year or three-year commitment and are usually a better fit for stable, predictable workloads.
Can Google Cloud discounts stack?
SUDs, CUDs, and Spot VMs do not stack with each other in any scenario. Usage can only be covered by one type of discount at any time. That means you need to evaluate which option delivers the best outcome for each workload instead of expecting multiple discounts to apply together.
When should I use Spot VMs?
Use Spot VMs when the workload can tolerate interruption and recovery is built into the architecture. Spot VMs are significantly cheaper than standard On-Demand VMs because they rely on Google Cloud’s surplus capacity. They are often a strong fit for batch jobs, data processing, and other non-critical workloads where lower cost matters more than guaranteed availability.
