Desktop and Application Streaming
Accelerate Cloud Development with Docker Desktop on Amazon WorkSpaces
Development teams want cloud desktops that feel like local machines. Until now, that meant giving up containers. Today, Amazon WorkSpaces Personal supports nested virtualization. All WorkSpaces Personal and Core bundles, including Custom, support nested virtualization except on Value bundles, Amazon Linux 2, Windows Server 2016, or Windows 10. For the best experience, we recommend enabling it on Power, PowerPro, and General Purpose bundles, providing full Docker Desktop functionality, including Windows Subsystem for Linux 2 (WSL2), Kubernetes, Docker Compose, and Android emulators.
For Standard and Performance bundles, Docker Offload provides full container access without local virtualization on the endpoint. Docker Offload is Docker Desktop running on a remote virtual machine (VM). The workflow and user experience remain the same. You build, run, and debug containers exactly as you would locally, while the compute happens remotely. The Amazon WorkSpace stays light, cost-efficient, and easy to provision.
These two approaches are complementary:
- Docker Desktop with nested virtualization — for developers whose work depends on native inner-loop performance, local Kubernetes clusters, Android emulators, and WSL2 toolchains.
- Docker Offload — for teams that prioritize portability, elastic scale, and cost efficiency, such as roaming developers, contractors, and data or machine learning (ML) workloads.
Both run on the same underlying platform. IT teams do not have to choose one strategy for the entire fleet. Platform engineers, full-stack developers, and mobile developers each get the container workflows their work demands, like Docker Compose, Docker-in-Docker, and emulator support, inside a managed, secure cloud workstation.
What Held Developers Back
Cloud desktops solve three hard problems: keeping source code off endpoints, eliminating environment drift with golden images, and provisioning developer machines in minutes instead of weeks. Containers were the gap. Docker Desktop requires a hypervisor. Cloud desktops could not provide one. Amazon WorkSpaces Personal with nested virtualization closes that gap.
One Platform, Two Container Strategies
Some developers need raw local performance. Others need portability and low cost. This launch gives both groups a native container path on Amazon WorkSpaces.
- Docker Desktop — uses nested virtualization to run containers locally. You get the full Docker experience with low-latency feedback, available on Power, PowerPro, and General Purpose bundles.
- Docker Offload — runs containers on a dedicated remote VM per developer—no queues, no shared resources. Your desktop stays light and portable. It is available on Standard and Performance bundles.
Deploy both strategies side by side in the same environment.
What You Can Build
This launch unlocks container workflows across every developer profile on Amazon WorkSpaces:
- Full-stack development — Run containers locally with Docker Compose or offload them remotely for a lighter desktop. No remote build queues in either case.
- Platform engineering — Spin up local Kubernetes clusters, run Docker-in-Docker, and debug CI pipelines inside the WorkSpace.
- Mobile development — Launch Android emulators alongside Docker Desktop. Build, test, and validate without separate hardware.
- WSL2 workflows — Full Linux toolchains, containers, and package managers on a managed Windows desktop.
- Roaming developers — Docker Offload on Standard bundles. Containers run remotely. The desktop stays light and portable.
- Contractors — Onboard in minutes on a Standard bundle with Docker Offload. When the project ends, deprovision the WorkSpace. No hardware to reclaim.
- Data and ML engineers — Offload training pipelines and heavy compute remotely. The desktop stays optimized for notebooks and IDEs.
Quick Reference
Match the developer to the bundle.
| Developer Scenario | Bundle | Solution | Primary Benefit |
| Full-stack remote dev | PowerPro | Docker Desktop | Low-latency inner loop |
| Platform / DevOps engineer | PowerPro | Docker Desktop | K8s, DinD, pipeline debugging |
| Mobile dev (Android) | PowerPro | Docker Desktop | Emulator needs nested virt |
| WSL2 + containers | PowerPro | Docker Desktop | WSL2 needs nested virt |
| Hybrid / roaming dev | Standard | Docker Offload | Cost efficiency + portability |
| Data / ML engineer | Standard / Perf | Docker Offload | Offload heavy compute |
| Contractor (cost-sensitive) | Standard | Docker Offload | Minimal bundle + on-demand containers |
Design for Cloud-Native, Not VDI
Good cloud developer environments share five traits:
- Fast provisioning — minutes, not weeks.
- Security by default — no code on endpoints, federated identity, AWS CloudTrail logging.
- Right-sized cost — PowerPro where needed, Standard where not.
- Resilience — multi-Availability Zone, repeatable golden images.
- Flexibility — change bundles without rebuilding; run both Docker strategies side by side.
The anti-pattern is treating WorkSpaces as VDI lift-and-shift. That approach locks in legacy constraints and misses what makes WorkSpaces architecturally valuable. Design for elasticity and developer self-service from day one.
Conclusion
With Amazon WorkSpaces and Docker, your developers have multiple paths to containerized development. Choose the approach that fits your team and start shipping faster. For more on Docker Offload, see the official GA announcement.
Authors
Noah Jackson – Specialist Solution Architect for Agentic WorkSpaces at Amazon Web Services, supporting customers across the Automotive & Manufacturing & Energy industries globally. With more than 20 years of experience in end-user computing and infrastructure, Noah specializes in helping organizations modernize their desktop and application environments using AWS. He has spent four years at AWS and focuses on technologies including Amazon WorkSpaces, Amazon WorkSpaces Applications, virtualization, identity, and cloud-based EUC solutions.