How Do You Change an EC2 Instance Type from t3.large to Compute Optimized?

SOA-C03 By islever Editorial Last updated Oct 20, 2026

Implement performance optimization strategies for compute, storage, and database resources. Provision and maintain cloud resources.

Short answer

C Stop the t3.large instance, change its instance type to a compute optimized type, then start it.

Question

A company has an Amazon EC2 instance that has high CPU utilization. The EC2 instance is a t3.large instance and is running a test web application. The company discovers that the web application would operate better on a compute optimized large instance.
What should a CloudOps engineer do to make this change?

Migrate the EC2 instance to a compute optimized instance by using AWS VM Import/Export.
Enable hibernation on the EC2 instance. Change the instance type to a compute optimized instance. Disable hibernation on the EC2 instance.
Stop the EC2 instance. Change the instance type to a compute optimized instance. Start the EC2 instance. Correct Answer
Change the instance type to a compute optimized instance while the EC2 instance is running.

Key Takeaways

The test is whether you know the instance-type change lifecycle for EBS-backed instances; the trap is assuming you can resize a running instance or that hibernation plays a role in the resize.

Changing an EC2 instance's type — such as moving a t3.large test web server to a compute optimized instance — requires stopping the instance, applying the new instance type, and starting it again. This page confirms why the stop-change-start sequence (C) is the only supported path and why VM Import/Export, hibernation and live resizing are not.

Common mistake: Many candidates pick option D, assuming a running EC2 instance can be resized in place like a container or Lambda memory setting; EC2 instance types are tied to the underlying host hardware and the instance must be stopped first.

Expert Analysis

Why the Answer Is Correct

EC2 instance types map to specific CPU, memory and network hardware, so the type can only be reassigned while the instance is not running on a host. For an EBS-backed instance, the supported workflow is exactly what option C describes: stop the instance, change the instance type (console, CLI modify-instance-attribute --instance-type, or API), then start it on new hardware. Because the root volume is EBS-backed, the data persists across the stop/start, and the elastic IP or public IP behavior is predictable (public IPv4 may change unless an Elastic IP is attached). Moving from t3.large to a compute optimized family such as c7i.large directly addresses the high CPU utilization the test web application is exhibiting. The learner comment from Bad_Mat reaches the same conclusion, stating "Stop the EC2 instance. Change the instance type to a compute optimized instance."

Why the Other Options Are Wrong

Option A is a category error: AWS VM Import/Export migrates VM images from on-premises hypervisors into EC2 as AMIs; it is not a mechanism for resizing an instance that already lives in AWS. Option B adds hibernation, which preserves RAM contents to EBS on stop and restores it on start — it does not remove the stop/start requirement, and hibernation has its own prerequisites (encrypted root volume, supported instance families, size limits). Option D is the classic trap: an instance type cannot be changed while the instance is in the running state; the console and API both reject the request until the instance is stopped. None of A, B or D can produce a compute optimized instance for a running t3.large.

Community Comment Notes

Bad_Mat's comment is short but aligns with the documented procedure, explicitly selecting the stop-change-start answer. The comment reinforces that the community treats the resize as a stop/start operation rather than an in-place or import operation. There is no dissent in the thread, so the consensus here matches both the wording of the question and AWS's documented behavior for EBS-backed instances.

Official Reference

These links open the official vendor documentation in a new tab.

Exam Strategy

When a stem mentions changing an instance's size, family or type, immediately look for the stop-change-start option and eliminate anything claiming a live resize. Remember that hibernation and VM Import/Export are different features, not shortcuts around the stop/start requirement.

Frequently Asked Questions

Can an EC2 instance type be changed while the instance is running?

No. The instance must be stopped (not terminated) before the type can be modified; the new type takes effect when you start it again.

Why doesn't enabling hibernation help change the instance type?

Hibernation only preserves RAM contents to the EBS root volume across a stop/start; the instance still must be stopped to change its type, so hibernation adds no benefit here.

More SOA-C03 FAQ →

Discussion 1 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 1

Bad_Mat 👍 1 Answered C

C. Stop the EC2 instance. Change the instance type to a compute optimized instance. Start the EC2 instance.

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Practice All SOA-C03 Questions

Access 100 questions with complete answers and detailed explanations.

View Full SOA-C03 Practice Test →

← Back to SOA-C03 Study Guide