openshift · rohennes · Jul 9, 2026
diff --git a/images/vcp-networking-arch-draft2.png b/images/vcp-networking-arch-draft2.png
diff --git a/images/vcp_networking_arch.png b/images/vcp_networking_arch.png
diff --git a/images/vcp_storage_arch.png b/images/vcp_storage_arch.png
diff --git a/modules/con_virt-vcp-architecture.adoc b/modules/con_virt-vcp-architecture.adoc
@@ -15,4 +15,6 @@ The target cluster's worker nodes run on separate infrastructure, either physica
 
 For high availability, distribute control plane VMs across different physical nodes on the hosting cluster.
 This anti-affinity placement ensures that if a physical node fails, only one control plane VM is affected and the remaining nodes maintain etcd quorum.
-Configure anti-affinity using pod anti-affinity rules or topology spread constraints in the VM specifications.
+Configure anti-affinity by using pod anti-affinity rules or topology spread constraints in the VM specifications.
+
+image::images/vcp-networking-arch-draft2.png[Virtualized control plane architecture]
diff --git a/modules/con_virt-vcp-storage-requirements.adoc b/modules/con_virt-vcp-storage-requirements.adoc
@@ -18,3 +18,5 @@ Consider the following when choosing a storage solution:
 
 * {odf-short} replicates data across multiple storage nodes for durability. This replication can cause unpredictable latency spikes during synchronization, which might affect `etcd` performance. Monitor `etcd` disk latency in production and ensure the underlying storage nodes use NVMe disks.
 * {lvms} writes directly to local disks without replication, providing consistent low latency. However, data is not replicated across nodes.
+
+image::images/vcp_storage_arch.png[Virtualized control plane architecture]
Original file line number	Diff line number	Diff line change
Expand Up		@@ -18,3 +18,5 @@ Consider the following when choosing a storage solution:

		* {odf-short} replicates data across multiple storage nodes for durability. This replication can cause unpredictable latency spikes during synchronization, which might affect `etcd` performance. Monitor `etcd` disk latency in production and ensure the underlying storage nodes use NVMe disks.
		* {lvms} writes directly to local disks without replication, providing consistent low latency. However, data is not replicated across nodes.

		image::images/vcp_storage_arch.png[Virtualized control plane architecture]