3.1.2.2 Stretched Resource Group
This less frequently used design stretches a resource group across both availability zones, with some clusters in the first site and others in the second. Resource group management is located in the primary site with failover to the second site.
The main advantage of this design is that it is possible to stretch organization networks across both sites and achieve Layer 2 adjacency for customer workloads. Disadvantages are lack of scalability, the need for disaster recovery of resource group management components, and the possibility of logical corruption of the management components that can down both availability zones.
Figure 3. Stretched Resource Group
Note None of these multisite designs requires stretched storage (storage metro cluster solution). However, stretched storage can be used for disaster recovery protection of the management layer. In such a case, distance between sites is limited by the supported round-trip latency of the storage network (usually 5–10 ms).
vCloud Director 8.20 and its components support a maximum 40 ms of network round-trip latency between sites (approximately 3300 km). However, the user experience diminishes with larger latencies and slower bandwidth.
