Add nuance around stretched clusters

docs/reference/high-availability/cluster-design.asciidoc

@@ -230,24 +230,47 @@ The cluster will be resilient to the loss of any node as long as:
 [[high-availability-cluster-design-large-clusters]]
 === Resilience in larger clusters
 
-It is not unusual for nodes to share some common infrastructure, such as a power
-supply or network router. If so, you should plan for the failure of this
+It is not unusual for nodes to share some common infrastructure, such as network
+interconnects or a power supply. If so, you should plan for the failure of this
 infrastructure and ensure that such a failure would not affect too many of your
 nodes. It is common practice to group all the nodes sharing some infrastructure
 into _zones_ and to plan for the failure of any whole zone at once.
 
-Your cluster’s zones should all be contained within a single data centre. {es}
-expects its node-to-node connections to be reliable and have low latency and
-high bandwidth. Connections between data centres typically do not meet these
-expectations. Although {es} will behave correctly on an unreliable or slow
-network, it will not necessarily behave optimally. It may take a considerable
-length of time for a cluster to fully recover from a network partition since it
-must resynchronize any missing data and rebalance the cluster once the
-partition heals. If you want your data to be available in multiple data centres,
-deploy a separate cluster in each data centre and use
-<<modules-cross-cluster-search,{ccs}>> or <<xpack-ccr,{ccr}>> to link the
+{es} expects its node-to-node connections to be reliable and to have low
+latency and adequate bandwidth. Many of the tasks that {es} performs require
+multiple round-trips between nodes. This means that a slow or unreliable
+interconnect may have a significant effect on the performance and stability of
+your cluster. A few milliseconds of latency added to each round-trip can
+quickly accumulate into a noticeable performance penalty. {es} will
+automatically recover from a network partition as quickly as it can but your
+cluster may be partly unavailable during a partition and will need to spend
+time and resources to resynchronize any missing data and rebalance itself once
+the partition heals. Recovering from a failure may involve copying a large
+amount of data between nodes so the recovery time is often determined by the
+available bandwidth.
+
+If you have divided your cluster into zones then typically the network
+connections within each zone are of higher quality than the connections between
+the zones. You must make sure that the network connections between zones are of
+sufficiently high quality. You will see the best results by locating all your
+zones within a single data center with each zone having its own independent
+power supply and other supporting infrastructure. You can also _stretch_ your
+cluster across nearby data centers as long as the network interconnection
+between each pair of data centers is good enough.
+
+[[high-availability-cluster-design-min-network-perf]] There is no specific
+minimum network performance required to run a healthy {es} cluster. In theory a
+cluster will work correctly even if the round-trip latency between nodes is
+several hundred milliseconds. In practice if your network is that slow then the
+cluster performance will be very poor. In addition, slow networks are often
+unreliable enough to cause network partitions that will lead to periods of
+unavailability.
+
+If you want your data to be available in multiple data centers that are further
+apart or not well connected, deploy a separate cluster in each data center and
+use <<modules-cross-cluster-search,{ccs}>> or <<xpack-ccr,{ccr}>> to link the
 clusters together. These features are designed to perform well even if the
-cluster-to-cluster connections are less reliable or slower than the network
+cluster-to-cluster connections are less reliable or performant than the network
 within each cluster.
 
 After losing a whole zone's worth of nodes, a properly-designed cluster may be