Skip to main content

Cost Saving Recommendations Overview

The nOps Recommendation engine helps you identify cost-saving opportunities across your cloud infrastructure. Our intelligent system continuously analyzes your environment to provide actionable recommendations that can help reduce waste and optimize your cloud spend.

Overview​

Recommendations are generated based on your cloud usage patterns, resource configurations, and industry best practices. Each recommendation includes:

  • Estimated cost savings
  • Implementation effort level
  • How-to-implement steps powered by Clara
  • Clara generated explanation of the issue, influenced by actual resource details

Clara Insights​

Clara Insights is an AI-powered feature that automatically analyzes your recommendations and highlights the most impactful cost optimization opportunities. Located at the top of the Recommendations page, the insights carousel displays clickable cards that help you quickly identify where to focus your optimization efforts.

How Clara Insights Works​

  • Automatic Analysis: Clara continuously analyzes your recommendation data to identify patterns by category, effort level, and savings potential
  • Clickable Insight Cards: Each insight card can be clicked to instantly filter and view matching recommendations
  • Cross-Platform Visibility: Weekly Recommendation Insights are also available on the Clara Feed page, so you can stay informed without navigating to the Recommendations page

Example Insights​

Clara generates insights such as:

  • "You have 12 high-impact Compute recommendations worth $5,000/month in potential savings"
  • "3 low-effort Storage optimizations can save $1,200/month with minimal implementation work"

Savings Achieved​

The Savings Achieved card tracks your realized savings from implemented recommendations. Located in the top summary section alongside the Savings Summary and Recommendations by Effort cards, it provides instant visibility into your optimization ROI.

Metrics Displayed​

  • Monthly Savings: Your current monthly savings rate from all implemented optimizations
  • Saved to Date: Cumulative total savings across your entire optimization history

As recommendations are marked as resolved (implemented), the system automatically calculates realized savings based on the estimated monthly savings for each recommendation.

Savings History​

The Savings History timeline provides a visual representation of your recommendation activity and savings achievements over time. This interactive chart helps you understand trends and demonstrate the impact of your cost optimization program.

Features​

  • Two View Modes:

    • Spend View: Visualizes Savings to Date, Cumulative Monthly Savings, and Realized Monthly Savings
    • Count View: Shows New Recommendations, Resolved Recommendations, and Total Active over time

  • Configurable Time Ranges: 7 days, 30 days, 90 days, 6 months, 1 year, or All time

  • Interactive Tooltips: Hover over any data point to see detailed information including savings achieved when recommendations were resolved

Recommendation Types

Upgrade EBS Volume​

Upgrading your EBS volume to a more efficient storage type (e.g., GP3 from GP2) can improve performance while reducing costs by taking advantage of lower pricing and better throughput.

EC2 Instances for Rightsizing​

Analyze your EC2 instance usage and resize underutilized instances to a more cost-effective instance type, ensuring that you maintain performance while reducing unnecessary expenses.
Effort Level: This recommendation can have different effort levels depending on the complexity of resizing. It may range from medium to high effort based on the need for validation, downtime, or application reconfiguration.

  • Medium Effort: Resizing across instance families (e.g., m5.large to c5.large) may require validation of application compatibility, as different families are optimized for different workloads (e.g., compute, memory, or storage). This might involve some downtime or reconfiguration.
  • High Effort: Resizing that requires significant changes, such as moving to a different region, changing the operating system, or reconfiguring the application to work with a new instance type. This could also include scenarios where downtime is unavoidable, or extensive testing is required to ensure compatibility.

Stop RDS Instance​

Identify and stop idle or underutilized RDS instances to eliminate costs associated with compute and storage resources that are not actively contributing to workloads.

Rightsize Lambda Function​

Optimize Lambda function memory and execution time to avoid over-provisioning, ensuring you only pay for the resources necessary to run your workloads efficiently.

Scale-In EC2 Auto Scaling Group​

Reduce the number of instances in an EC2 Auto Scaling Group when demand is low to avoid paying for unused capacity while maintaining the ability to scale up when needed.

Migrate EC2 to Graviton​

Transition EC2 instances from x86-based processors to AWS Graviton instances, which offer better price-performance ratios, leading to lower compute costs for compatible workloads.

Migrate RDS to Graviton​

Move RDS databases to Graviton-powered instances to achieve improved performance and lower operational costs while maintaining compatibility with popular database engines.

Upgrade EC2 Instance​

Upgrade your EC2 instances to a newer generation within the same family or a more optimized instance type to gain better performance at a lower or similar cost.

Migrate Auto Scaling Group to Graviton​

Convert Auto Scaling Groups running x86-based instances to Graviton-based instances to improve efficiency and reduce costs while maintaining scalability.

Unused EBS Volume​

Identify and delete EBS volumes that are detached or unused to eliminate unnecessary storage costs and optimize resource allocation.

Upgrade EC2 Auto Scaling​

Modify EC2 Auto Scaling Groups to use newer, more cost-effective instance types or instance purchase options (e.g., Spot Instances) to reduce compute costs while maintaining availability.

Delete Active EC2 Snapshots Older than 90 Days​

Identify and delete active EC2 snapshots that are older than 90 days to reduce storage costs associated with outdated backups while ensuring compliance with your data retention policies.

Delete High-Cost RDS Snapshots​

Identify and delete high-cost RDS snapshots that are no longer needed to optimize storage expenses and free up resources for more critical workloads.

Aurora IO Optimization​

Heavily reduce Aurora costs by switching to IO Optimized when the usage pattern indicates heavy IO usage. Effort Level: This recommendation can range from low to high effort depending on the following factors:

  • Low Effort: When neither extended support nor a database upgrade is needed.
  • Medium Effort: When extended support is required but no database upgrade is needed.
  • High Effort: When a database upgrade is required, regardless of whether extended support is needed.

S3 Intelligent Tiering​

Enable S3 Intelligent Tiering to automatically move data between storage tiers based on access patterns, reducing costs for infrequently accessed data without manual intervention.
Effort Level: Typically low, as this involves enabling a feature with minimal configuration changes.

AI Model Recommendation​

Identify opportunities to optimize AI model usage by recommending transitions from OpenAI models to AWS Bedrock or other cloud-native AI services, potentially reducing costs while maintaining performance and capabilities.

ELB Recommendation​

Identify unused or underutilized Elastic Load Balancers (ELBs) that can be removed to eliminate unnecessary costs while maintaining application availability and performance for active workloads. Effort Level: Typically low to medium, depending on:

  • Low Effort: When the ELB is clearly unused with no active connections or traffic for an extended period.
  • Medium Effort: When the ELB has minimal traffic but requires validation to ensure it's not supporting critical but infrequent workloads.

Effort Level​

Determining the effort level of a recommendation involves evaluating several key factors. One major consideration is whether the recommendation requires validation from engineering teams, as this can introduce additional coordination and review processes. Another critical factor is whether implementing the recommendation necessitates restarting a resource, which could lead to temporary downtime or service disruptions. Additionally, the complexity of implementation is influenced by whether changes to operating system parameters, reinstallation of applications, or updates to application configurations are required. In cases where a recommendation involves rewriting parts of the application, the effort level can increase significantly, as this may require code modifications, testing, and deployment. Other aspects, such as potential impacts on system performance, compatibility with existing infrastructure, and the need for extensive documentation or user training, may also contribute to the overall effort required for implementation.

FAQs​

Expand FAQs

1. Who is this for?​

The Recommendations feature is available for all Inform customers, including both trial and paying customers.

2. What is Clara Insights?​

Clara Insights is an AI-powered feature that automatically analyzes your recommendations and surfaces the most impactful cost optimization opportunities. The insights appear as clickable cards at the top of the Recommendations page, and Weekly Recommendation Insights are also available on the Clara Feed page.

3. How is Savings Achieved calculated?​

Savings Achieved tracks the estimated monthly savings from recommendations that have been marked as resolved (implemented). As you implement recommendations and mark them complete, the system calculates your realized savings based on the estimated monthly savings for each recommendation. These values accumulate over time to show your total savings to date.

4. How are recommendations generated?​

Our system analyzes your cloud infrastructure data to identify optimization opportunities. We use a combination of:

  • Resource utilization patterns
  • AWS best practices
  • Industry benchmarks
  • Historical usage data

Recommendations are refreshed regularly to ensure you always have the most up-to-date information.

5. What types of recommendations are provided?​

We provide various types of recommendations, including:

  • Rightsizing opportunities: Identify over-provisioned resources that can be downsized
  • Idle resources: Detect unused or underutilized resources that can be terminated
  • Reserved Instance opportunities: Suggest Reserved Instance purchases for consistent workloads
  • Storage optimizations: Identify storage that can be moved to lower-cost tiers
  • Modernization suggestions: Recommend newer, more cost-effective services

6. Can I ignore or dismiss recommendations?​

Yes, you can override any recommendation if you decide not to implement it. This helps tailor the system to your specific needs and prevents the same recommendation from appearing repeatedly.

To override a recommendation:

  1. Open the recommendation details
  2. Click the "Override" button

7. How accurate are the estimated savings?​

Savings estimates are based on your current usage patterns and AWS pricing. While we strive for accuracy, actual savings may vary based on:

  • Changes in your usage patterns
  • AWS price changes
  • Implementation specifics

8. Can I share recommendations with my team?​

Yes, you can share recommendations in several ways:

  • Create and share custom recommendation reports
  • Schedule regular email notifications with recommendation summaries
  • Export recommendations to CSV for further analysis

9. How often are recommendations updated?​

Recommendations are updated daily to reflect the latest data from your cloud environment. This ensures you always have access to the most current optimization opportunities.

10. How do I prioritize which recommendations to implement first?​

We recommend prioritizing based on:

  1. Highest potential savings
  2. Lowest implementation effort
  3. Lowest operational risk

The recommendations dashboard allows you to sort and filter by these criteria to help with prioritization. You can also use Clara Insights to quickly identify the most impactful opportunities.

11. Where can I see Weekly Recommendation Insights?​

Weekly Recommendation Insights appear on both the Recommendations page (via Clara Insights) and the Clara Feed page. This cross-platform visibility ensures you never miss important savings opportunities, regardless of where you are in nOps.