AWS Transform MGN (formerly AWS Application Migration Service) vs Google Compute Engine
AWS Transform MGN (formerly AWS Application Migration Service)
aws.amazon.com
Google Compute Engine
cloud.google.comSide by side
| Figure | AWS Transform MGN (formerly AWS Application Migration Service) | Google Compute Engine |
|---|---|---|
| Price from | Free | Free |
| Free plan | Yes | Yes |
| Free trial | — | Yes |
| Open source | — | — |
| Web | Yes | Yes |
| Windows | Yes | Yes |
| Linux | Yes | Yes |
| API | Yes | Yes |
| Plans published | 1 | 2 |
| Spec score · Cloud Migration Tools | 8.2 (#4) | 7.8 (#12) |
| Availability | AWS says MGN is available in commercial AWS Regions and both AWS GovCloud (US) Regions. | The product page states single-instance availability SLAs of 99.95% for memory-optimized VMs and 99.9% for other VM families. |
| Compliance | AWS says MGN retains certifications including FedRAMP High, HIPAA, PCI DSS, ISO, and SOC 1, 2, and 3. | Google Cloud lists certifications and attestations including ISO/IEC 27001, ISO/IEC 27017, ISO/IEC 27018, SOC 1/2/3, and PCI DSS. |
| Security | MGN requires TLS 1.2 or later, uses TLS and IAM for authentication, and recommends EBS encryption and limiting replication server exposure to the public internet. | Compute Engine offers Confidential VMs to encrypt data in use and Shielded VMs to defend against rootkits and bootkits. |
| Support | AWS documentation says support for installation or replication issues is limited to operating systems listed as supported. | Every Google Cloud customer, including Free Trial users, receives Basic Support with documentation, community forums, billing assistance, and Active Assist. |
Plans
AWS Transform MGN (formerly AWS Application Migration Service)
- Free periodFree
AWS Transform MGN (formerly AWS Application Migration Service) pricing →
Google Compute Engine
- Compute Engine Free TierFree
- Pay-as-you-go Compute EnginePrice not published
In their own words
AWS Transform MGN (formerly AWS Application Migration Service)
AWS Transform MGN automates rehosting physical, virtual, and cloud servers as native Amazon EC2 instances. It continuously replicates source-server data at the block level, converts servers for AWS, and coordinates cutover with minimal downtime. Teams can use an agentic AWS Transform workflow or manage migration steps in the MGN console, switching paths without losing progress. Sources include physical servers, VMware vSphere, Microsoft Hyper-V, other on-premises infrastructure, public clouds, and migrations across AWS Regions, Availability Zones, or accounts. After launch, built-in or custom actions can configure disaster recovery, upgrade operating systems, convert licenses, or install CloudWatch Agent. Servers can be organized into applications and waves for launch, cutover, archival, and configuration operations. MGN supports x86 sources and replication volumes up to 16 TB; larger volumes are not supported. A 90-day free period covers 2,160 hours per source server when used continuously. AWS infrastructure for replication, testing, or cutover is charged separately, and continued replication is charged per hour after the free period.
Google Compute Engine
Google Compute Engine is an infrastructure-as-a-service product for running self-managed virtual machines on Google infrastructure. Choose a predefined machine type or configure a custom VM with 1 to 96 vCPUs and up to 8.0 GB of memory per core. Linux and Windows VMs are supported, with images including Debian, Ubuntu, FreeBSD, and Windows Server. Storage options include Persistent Disk and Local SSD. Managed instance groups can add VMs for peak load and replace underperforming instances; live migration can move a VM to another host in the same zone when its host goes down. Confidential VMs encrypt data in use, while Shielded VMs defend against rootkits and bootkits. The service integrates with Cloud Storage, App Engine, and BigQuery, and resources can be controlled through a REST-based API. Its free tier is free, within monthly limits and eligible US regions. Pay-as-you-go pricing is usage-based, with VM resources billed after a one-minute minimum in one-second increments.