ASG Rightsizing
Rightsizing ASG fleets means analyzing every instance, picking the right type, updating the launch template, and rolling it out without downtime. Infralyst handles all of it in a single Terraform PR.
Unlimited free recommendations · Read-only IAM role · Your team reviews and merges every change
| ASG | CPU | Memory | Current | Recommended | Savings | |
|---|---|---|---|---|---|---|
| web-api-produs-east-1 · 84 days | 5.1%p99.5 21% | 9%p99.5 18% | c5.2xlarge | c5.xlarge | $94/mo | Generate PR |
| worker-queueus-east-1 · 60 days | 7.3%p99.5 19% | 11%p99.5 23% | m5.xlarge | m5.large | $47/mo | Generate PR |
| cache-clusterus-east-1 · 45 days | 6.2%p99.5 17% | 14%p99.5 28% | r5.large | r5.medium | $36/mo | Generate PR |
Your launch templates are set
and never revisited
Teams pick an instance type when they first create an ASG, usually sized for anticipated peak load. Workloads change, utilization drops, but the launch template doesn't. Nobody wants to touch it because changing the instance type means coordinating a fleet-wide instance refresh without dropping traffic. So the waste compounds quietly.
AWS Cost Explorer will flag the opportunity. But it stops at "you could save money here." Infralyst does the rest. It analyzes real utilization across every instance in the ASG, selects a conservative target, and generates a Terraform PR that updates your launch template and adds an instance refresh block for zero-downtime rollout.
How the analysis works
A conservative, multi-gate approach. If it's not safe to ship, it doesn't get recommended.
1. Collect utilization metrics
Infralyst pulls CPU and memory utilization from CloudWatch for every instance in the ASG, using up to 365 days of historical data.
2. Run the analysis gates
Each ASG passes through data quality validation, utilization thresholds, and seasonality checks. Every instance in the fleet has to clear the bar. An ASG only gets a recommendation if all gates pass.
3. Generate the recommendation
Infralyst recommends the next size down in the same instance family. You review the recommendation and generate a PR when you're ready. The PR updates the launch template and adds an instance refresh block for a zero-downtime rolling replacement.
Thresholds & methodology
Every gate and threshold used to determine whether an ASG is safe to downsize. These thresholds are intentionally conservative. We'd rather miss a savings opportunity than recommend a change that causes problems.
Pre-flight
Data quality
CPU & Memory
Seasonality
For ASGs with 60–365 days of data, P95 is checked per weekly bucket. If any week exceeds the 40% threshold, the ASG is excluded.
Infralyst only recommends stepping down one instance size at a time. If there's room to go further, it happens across multiple analysis cycles.
What the PR looks like
Infralyst opens a Terraform pull request in your repo. Your team reviews and merges it like any other change.
Downsize ASG web-api-prod: c5.2xlarge → c5.xlarge (-$94/mo)
Auto Scaling Group web-api-prod in us-east-1 can be downsized from c5.2xlarge to c5.xlarge, saving an estimated $94/mo.
| Metric | Avg | P99.5 |
|---|---|---|
| CPU | 5.1% | 21% |
| Memory | 9% | 18% |
Lookback: 84 days
Trigger PRs directly from Slack
Get notified when new savings are found. Review the details and generate a PR without leaving Slack.
New ASG savings recommendation · Save $94.00/mo
ASG
web-api-prod (us-east-1)
Lookback
84d
CPU Avg
5.1%
CPU P99.5
21%
Mem Avg
9%
Mem P99.5
18%
Current
c5.2xlargeRecommended
c5.xlargeView on Infralyst
Safety & guardrails
Designed to be safe to run in production. Here's what Infralyst won't do.
Conservative thresholds
Infralyst won't recommend a downsize unless CPU and memory P99.5 (or P95 for longer windows) are comfortably below 40% across the ASG fleet. A single spike won't trigger a false recommendation.
Seasonality-aware
For ASGs with 60+ days of data, P95 utilization is checked per weekly bucket over the full observation window. If any week exceeds 40%, the ASG is excluded. No recommendation, no risk.
Zero-downtime instance refresh
Every PR includes an instance refresh block with a calculated min_healthy_percentage. AWS replaces instances one batch at a time so your fleet stays up during the rollout. Single-instance ASGs get a downtime warning in the PR instead.
PR-only changes
Infralyst never applies changes directly. Every recommendation is delivered as a notification in Slack or in our console. You trigger PR generation, review, test, and merge on your timeline.
Also available for EC2 Instances.
Learn more about ASG optimization
How to Enable EC2 Memory Metrics in CloudWatch
A simple guide to getting RAM usage data into CloudWatch so you can actually perform EC2 rightsizing without flying blind.
March 27, 2026
Best Practices for EC2 Rightsizing
A practical guide to EC2 rightsizing within the same instance family: how much data you need, what thresholds to use, and the guardrails that keep you out of trouble.
December 19, 2025Want more? View all articles
Start rightsizing your Auto Scaling Groups
Connect your AWS account and Infralyst will find ASGs you can safely downsize. Generate a Terraform PR that updates the launch template and handles the instance refresh, when you're ready.
Start free with 3 PRsNo credit card required · Read-only IAM role · Your team reviews and merges every change