Blog posts
Events
Cloudgov University
As cloud adoption grows, effective cost management is critical for organizations. AWS, Azure, and Google Cloud Platform (GCP) each offer savings mechanisms such as Savings Plans and Reserved Instances (RIs) to help customers optimize cloud costs. This guide provides a comprehensive comparison of these options, using a customer example with $75,000 in monthly compute spend and $25,000 in database spend. Additionally, it includes steps for estimating Savings Plans or RIs.
Understanding Savings Mechanisms Across Cloud Providers
Each cloud provider offers distinct programs that provide discounts for committed usage over specific periods. Here’s a comparison of the offerings from AWS, Azure, and GCP.
AWS: Savings Plans and Reserved Instances
AWS provides two primary cost-saving options: Savings Plans and Reserved Instances (RIs).
Savings Plans: Flexible pricing models offering significant discounts for a commitment to consistent usage ($/hour) over one or three years.
- Compute Savings Plans: Offer up to 66% savings and apply to EC2, Fargate, and Lambda usage, regardless of instance type, OS, tenancy, or region. This allows flexibility across services without losing discounts.
- EC2 Instance Savings Plans: Provide up to 72% savings on specific instance families and regions, with the flexibility to adjust instance sizes within the family.
Reserved Instances (RIs): Allow customers to save up to 72% by committing to specific instance configurations for one or three years. There are two types:
- Standard RIs: Provide the highest discount but with limited flexibility to modify attributes.
- Convertible RIs: Allow modifications to instance attributes like family, OS, and tenancy, offering more flexibility but at a slightly reduced discount.
| Feature | Compute Savings Plans | EC2 Instance Savings Plans | Reserved Instances |
| Savings Potential | Up to 66% | Up to 72% | Up to 72% |
| Applies To | EC2, Fargate, Lambda | EC2, Fargate, Lambda EC2 instances within a specified instance family/region | EC2, RDS, Elastic Cache, OpenSearch |
| Flexibility | High – change family, OS, and region anytime |
Medium – restricted to family/region but size is flexible |
Low – restricted to specific instance configuration |
| Ideal For | Dynamic, fluctuating workloads | Slightly stable workloads | Highly predictable, stable workloads |
Estimating Savings with AWS
To manually calculate how much Savings Plans or RIs to purchase, customers can follow these steps:
Analyze Historical Usage with AWS Cost Explorer:
- Go to AWS Cost Explorer and review your last 3-6 months of usage.
- Use Savings Plans recommendations in Cost Explorer to estimate the ideal commitment level for consistent, predictable workloads.
Identify Consistent Workloads:
- Identify workloads with consistent usage patterns that can be committed to Reserved Instances (e.g., database instances on RDS).
Calculate Total Commitments:
- Based on average hourly usage, decide the dollar-per-hour commitment level for Compute Savings Plans or the specific configurations for RIs.
Pricing Source for AWS: https://aws.amazon.com/savingsplans/pricing/
Azure: Savings Plans and Reserved Instances
Azure offers Savings Plans for Compute and Reserved Instances as cost-saving options:
- Azure Savings Plans for Compute: Offer up to 65% savings on eligible compute services, such as VMs, dedicated hosts, and Azure Premium Functions, across instance families, sizes, OS, and regions.
- Reserved Instances (RIs): Allow savings of up to 72% on specific virtual machine configurations for one or three years. RIs are suitable for predictable, long-term workloads with fixed usage patterns.
| Feature | Azure Savings Plans for Compute | Azure Reserved Instances |
| Savings Potential | Up to 65% | Up to 72% |
| Applies To | Virtual machines, dedicated hosts, functions |
Specific VM configurations |
| Flexibility | High – change size, series, and region anytime |
Low – fixed to configuration |
| Ideal For | Dynamic workloads needing flexibility | Consistent, predictable workloads |
Estimating Savings with Azure
To manually calculate how much Azure Savings Plans or RIs to purchase, customers can follow these steps:
Use Azure Cost Management + Billing:
- Navigate to Azure Cost Management and Analyze historical compute usage over the last 3-6 months.
- Identify the average compute and database spend to determine predictable workloads suitable for RIs.
Review Recommendations in Azure Advisor:
- Go to Azure Advisor for personalized Savings Plan and RI recommendations.
- Azure Advisor provides insights based on usage trends to help estimate the optimal commitment level.
Estimate Commitments for Compute and Database Workloads:
- Determine the hourly usage and specific VM configurations based on Azure Advisor’s recommendations.
Pricing Source for Azure: https://azure.microsoft.com/en-us/pricing/reserved-vm-instances/
GCP: Committed Use Contracts (CUDs)
GCP’s Committed Use Contracts (CUDs) provide substantial discounts for customers who commit to specific usage levels.
- Resource-Based CUDs: Offer savings based on a committed amount of vCPUs and memory usage in a specific region, allowing flexibility within instance types.
- Spend-Based CUDs: Provide discounts based on a dollar-per-hour commitment across multiple regions and services, ideal for dynamic, multi-region workloads.
| Feature | Spend-Based CUDs | Resource-Based CUDs | Flexible CUDs |
| Savings Potential | Up to 45% | 40-60% | Varies based on specific usage |
| Applies To | Various compute resources, databases | Compute resources (vCPU, memory, storage) | Applicable across instance families/regions |
| Flexibility | High-spend-based, any instance family |
Medium-committed to region and family |
High-cross-region and family flexibility |
| Ideal For | Dynamic workloads across regions |
Predictable usage within a region |
Highly dynamic multi-region workloads |
Estimating Savings with GCP
To manually estimate how much CUDs to purchase, follow these steps:
Analyze Usage with Google Cloud Billing Reports:
- Go to Billing Reports and analyze usage over the last 3-6 months to identify consistent usage patterns.
Use Google Cloud’s Committed Use Analysis Tool:
- Access Committed Use Analysis within the Billing Console for personalized CUD recommendations based on your actual spend and usage.
- Review historical usage by region and by instance type to determine the appropriate level of commitment.
Set CUD Commitments Based on Consistent Workloads:
- Estimate monthly spend commitments for predictable workloads to cover stable compute and database services.
Pricing Source for GCP: https://cloud.google.com/compute/all-pricing
Customer Savings Optimization Example: Optimizing $75,000 Monthly Compute and $25,000 Monthly Database Spend
Using one-year, no-upfront payment options, here’s how a company with a monthly spend of $75,000 on compute and $25,000 on database services could optimize costs across AWS, Azure, and GCP:
| Cloud Provider | Monthly Compute Spend | Compute Savings (%) |
Monthly Database Spend | Database Savings (%) |
Total Monthly Savings |
New Monthly Spend |
| AWS | $75,000 | 45-55% | $25,000 | 30 – 40% | $41,250 – $51,250 | $48,750 – $58,750 |
| Azure | $75,000 | 45-60% | $25,000 | 35 – 45% | $42,500 – $56,250 | $43,750 – $57,500 |
| GCP | $75,000 | 40-55% | $25,000 | 35 – 45% | $38,750 – $52,500 | $47,500 – $61,250 |
Summary of Savings Recommendations
- AWS: Leverage Compute Savings Plans for dynamic compute workloads and RDS Reserved Instances for database services.
- Azure: Apply Savings Plans for Compute for flexibility in compute workloads, and RIs for Azure SQL Database for stable database usage.
- GCP: Utilize Spend-Based CUDs for compute resources, and Resource-Based CUDs for specific database configurations.
Typical Steps Summary Across Cloud Providers to figure out how to dollar amount savings plans or reservations to buy?
Here’s a consolidated checklist of manual steps for estimating Savings Plans or RIs:
Analyze Historical Usage:
- Use AWS Cost Explorer, Azure Cost Management, and GCP Billing Reports to review your last 3-6 months of cloud usage.
Review Recommendations:
- Leverage native recommendation tools like AWS Savings Plans Recommendations, Azure Advisor, and GCP’s Committed Use Analysis for personalized insights.
Calculate Commitments:
- Based on historical usage patterns and recommendation insights, estimate hourly or resource-based commitments for each cloud provider to maximize potential savings.
Why All Businesses Should Consider Savings Choices
Startups: Early-stage companies often have limited budgets. Savings plans and reserved instances can help them maximize their resources and extend their runway.
Mid-size organizations: As businesses grow, so do their cloud costs. Savings plans and reserved instances can help them manage expenses and invest in other growth areas.
Enterprises: Large enterprises with substantial cloud deployments can benefit significantly from the cost savings offered by these options, freeing up capital for strategic initiatives.
Considerations for Choosing the Right Option:
- Workload Predictability: For stable workloads, reserved instances typically offer higher discounts. For dynamic workloads, savings plans provide greater flexibility.
- Commitment Length: Longer commitments usually come with higher discounts but less flexibility.
- Payment Options: Consider upfront vs. monthly payment options based on your financial planning.
By understanding the various savings options and carefully evaluating their needs, businesses of all sizes can achieve significant cost reductions and optimize their cloud investments.
Conclusion
While each cloud provider offers tools for managing and optimizing cloud costs, calculating the optimal level of Savings Plans, Reserved Instances, or Committed Use Discounts manually can be complex and time-consuming and takes away precious time of your engineering teams from delivering features to looking through bills and utilization metrics. Cloudgov.ai empowers businesses of all sizes to maximize the value of their cloud investments through the world’s first Gen AI Agent FinOps platform. Our AI/ML technology enables executives, engineering, DevOps, Finance, and FinOps teams to make decisions based on Gen AI-driven actionable insights, realizing significant savings on multicloud and SaaS spend. With Cloudgov.ai, you can automate and optimize cost management across AWS, Azure, and GCP, ensuring maximum savings and value from your cloud investments.
Next Steps
Ready to unlock the full potential of your cloud investment? Start by signing up for a free trial and explore how Cloudgov.ai can transform your cloud cost management. Our Gen AI-powered platform offers precise, actionable insights to help you optimize spend across AWS, Azure, and GCP with ease. Whether you’re managing complex multicloud environments or seeking savings recommendations tailored to your workloads, Cloudgov.ai empowers you to make data-driven decisions, maximize savings, and drive efficiency. Take the first step toward smarter cloud spending today! Sign Up for 14 day free trial.
References
- AWS Savings Plans Pricing:https://aws.amazon.com/savingsplans/pricing/
- AWS Reserved Instances Pricing:https://aws.amazon.com/ec2/pricing/reserved-instances/
- Azure Reserved VM Instances Pricing:https://azure.microsoft.com/en-us/pricing/reserved-vm-instances/
- Google Cloud Compute Pricing: https://cloud.google.com/compute/all-pricing
- Google Cloud Committed Use Contracts Documentation: https://cloud.google.com/compute/docs/instances/signing-up-committed-use-contracts