Amazon WorkSpaces Cross-Region VDI Migration: Wolters Kluwer’s Windows 11 and Teams Lessons

The short version​

AWS reports that Wolters Kluwer (WK) moved its fleet from Asia Pacific (Singapore) to Asia Pacific (Mumbai) in waves, with automated provisioning and parallel running. According to AWS, round-trip latency fell from about 270 ms to 49 ms. That is roughly an 82% cut, and performance-related service desk calls fell by 40%. AWS says no workday was disrupted. The post gives no downtime figures, ticket baselines, sample sizes, measurement periods or costs.

Why they moved​

WK deployed WorkSpaces in Singapore in 2019. That was the nearest available Region to its mostly India-based staff at the time. As the user base grew to roughly 3,000 knowledge workers, Teams became central to daily work. Users reported:

• Teams audio and video out of sync
• Laggy screen sharing
• Sluggish application response
• Workarounds such as turning off video or avoiding screen share

AWS puts the root cause down to network latency from routing streams between India and Singapore. WorkSpaces had been available in Mumbai since 2021, but WK says a migration of this size needed a trigger.

The Teams optimization trigger​

The trigger was Microsoft Teams optimization for WorkSpaces, which reached general availability in February 2026. Microsoft's announcement is dated Feb 19, 2026. It describes the first VDI optimization for Microsoft Teams on Amazon WorkSpaces for Windows endpoints. AWS says the preview began in September 2025.

The mechanism is media offload. A Teams VDI bridge on the VM detects the virtual environment and opens a virtual channel to the endpoint. The WorkSpaces client then loads a Teams plugin, which pulls in a Slimcore MSIX package. A local process handles the media offload.

WK's logic is that once media runs locally, the remaining bottleneck is distance to the Region. These are AWS's descriptions, and the case study doesn't separate how much each factor contributed.

Check the scope before you copy this​

• Microsoft's announcement says the optimization is available for WorkSpaces accessed through the Windows WorkSpaces Client, and AppStream and other platforms are not yet supported.
• AWS's admin guide lists Teams 2.0 under Media Optimized RTC for DCV WorkSpaces only. The guide also warns that the client extension must be installed and separately updated. It adds that screen sharing can behave differently, and that client-side extensions may not suit BYOD or shared kiosk setups.
• The case study doesn't say which protocol or client fleet WK used. Don't assume your environment matches theirs.

How the migration worked​

AWS notes that each Region keeps its own directory services, custom bundles and user data. A cross-Region move is therefore a reprovisioning job, not a relocation. WK's goal was for users to log off a Singapore desktop on Friday evening and sign in to a configured Mumbai desktop on Monday morning.

WK also validated before committing. It provisioned a few Mumbai desktops and benchmarked them with Liquidware Stratusphere UX.

Wave structure​

1. A pilot of power users, including a volunteering director, validated the process end to end.
2. Each wave ran Singapore and Mumbai desktops in parallel for one to two weeks, so users could report missing configuration.
3. Later waves were sized for efficiency.
4. The Singapore desktop was decommissioned only after the user confirmed.

AWS says the pilot director asked for their whole team to be moved the following week.

The Windows 11 25H2 bundling​

Combining the Region move with a Windows 11 25H2 upgrade meant one disruptive event instead of two. That is sensible change management. The trade-off is that two changes are now coupled, so a problem could be harder to attribute and roll back. The case study doesn't establish that bundling suits every organization.

Microsoft's release page shows version context. Windows 11 25H2 became generally available on 2025-09-30, with support ending for Enterprise and Education editions on 2028-10-10. The case study doesn't state WK's edition, build number or servicing process. Admins planning something similar should confirm lifecycle dates for their own editions.

Reading the numbers with care​

• The 270 ms to 49 ms figure is WK's measured result for its users and routes. It isn't a general benchmark for Singapore versus Mumbai.
• The 40% drop in service desk calls has no stated baseline or time window.
• "Without disruption" is AWS's characterization and is not quantified.
• AWS publishes the post and sells WorkSpaces, and its closing invites readers to contact their account team. Treat it as a customer success story, not a neutral study.

A practical checklist​

1. Pilot in the target Region and measure it. Don't rely on map distance.
2. Inventory applications and per-user settings before building anything.
3. Find out where user data actually lives. Cloud-synced files are easy, while local D: drive data needs a defined window.
4. Automate app delivery, and use layering for awkward apps.
5. Keep the old desktops running through a validation period, with explicit sign-off before retirement.
6. If you adopt Teams optimization at the same time, confirm client OS, protocol, extension management and screen-sharing behavior first.
7. Weigh whether combining an OS upgrade with the move reduces risk or only concentrates it.

Bottom line​

The article's main value is the pattern: benchmark first, automate, run in parallel, migrate in waves and let users sign off. The migration's success was also helped by WK's data already living in OneDrive. Organizations with large local data or unmanaged apps will have more work to do.