Overview
When EC2 instance capacity is unavailable in the selected Availability Zone (AZ) or Region, Viz Now users have the option to move their workloads to an alternate location where resources are available.
This strategy can help maintain service continuity by avoiding congested zones or regions, especially when dealing with GPU-intensive or high-demand instance types.
Options for Relocating Workloads
✅ Move Entire Deployment
Viz Now does not currently support changing the Region or AZ for an existing space. To move your entire deployment, you must:
Create a new Viz Now space in the target Region and Availability Zone
Re-deploy all required applications in the new space
Manually reconfigure capacity applications and reservations.
🔄 Move a Single Application
You may also split your workload across locations:
Deploy a new Viz Now space in a different AZ or Region
Use the VPC Peering Tool to connect the new space with your existing one
Allow inter-space communication between apps (for example, using NDI Discovery)
⚠️ Note: When separating apps across zones or regions, ensure connectivity requirements are fully validated.
Considerations and Tradeoffs
Network Latency
Performance-sensitive apps (for example, live video, replay) may experience latency if apps are spread across distant locations
Delay settings may need to be adjusted.
Cost Implications
Intra-region AZ traffic may incur additional data transfer charges from AWS
Long distance Region-to-Region traffic is usually billed at a higher rate.
Best Practices
Monitor performance after migration
Use availability forecasts or historical insights from AWS to guide decisions - consider leveraging your AWS Technical Account Manager (TAM) to gain early visibility into capacity trends and regional constraints
Keep production-critical services together in the same location whenever possible.
Summary
Choosing an alternate AZ or Region is a valuable fallback strategy when managing scarce EC2 resources in Viz Now. With proper configuration and awareness of the tradeoffs, this can help ensure smooth, uninterrupted production workflows.