diff --git a/docs/operator-manual/disaster_recovery.md b/docs/operator-manual/disaster_recovery.md
new file mode 100644
index 0000000000000..fdc9018d76455
--- /dev/null
+++ b/docs/operator-manual/disaster_recovery.md
@@ -0,0 +1,25 @@
+# Disaster Recovery
+
+You can use `argocd-util` can be used to import and export all Argo CD data.
+
+Make sure you have `~/.kube/config` pointing to your Argo CD cluster.
+
+Figure out what version of Argo CD you're running:
+
+```bash
+argocd version | grep server
+# ...
+export VERSION=v1.0.1
+```
+
+Export to a backup:
+
+```bash
+docker run -v ~/.kube:/home/argocd/.kube --rm argoproj/argocd:$VERSION argocd-util export > backup.yaml
+```
+
+Import from a backup:
+
+```bash
+docker run -v ~/.kube:/home/argocd/.kube --rm argoproj/argocd:$VERSION argocd-util import - < backup.yaml
+```
diff --git a/docs/operator-manual/high_availability.md b/docs/operator-manual/high_availability.md
new file mode 100644
index 0000000000000..ad5106e72b903
--- /dev/null
+++ b/docs/operator-manual/high_availability.md
@@ -0,0 +1,19 @@
+# High Availability
+
+Argo CD is largely stateless, all data is persisted as Kubernetes objects, which in turn is stored in Kubernetes' etcd. Redis is only used as a throw-away cache and can be lost. When lost, it will be rebuilt without loss of service.
+
+A set HA of manifests are provided for users who wish to run Argo CD in a highly available manner. This runs more containers, and run Redis in HA mode.
+
+[Manifests ⧉](https://github.com/argoproj/argo-cd/tree/master/manifests) 
+
+!!! note
+    The HA installation will require at least three different nodes due to pod anti-affinity roles in the specs.
+ 
+## Scaling Up
+
+You might scale up some Argo CD services in the following circumstances:
+
+* The `argocd-repo-server` can scale up when there is too much contention on a single git repo (e.g. many apps defined in a single git repo).
+* The `argocd-server` can scale up to support more front-end load.
+
+All other services should run with their pre-determined number of replicas. The `argocd-application-controller` must not be increased because multiple controllers will fight. The `argocd-dex-server` uses an in-memory database, and two or more instances would have inconsistent data. `argocd-redis` is pre-configured with the understanding of only three total redis servers/sentinels.
diff --git a/mkdocs.yml b/mkdocs.yml
index af61cb7d388da..6f502141eff32 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -30,6 +30,8 @@ nav:
     - operator-manual/rbac.md
     - operator-manual/security.md
     - operator-manual/cluster-bootstrapping.md
+    - operator-manual/high_availability.md
+    - operator-manual/disaster_recovery.md
     - operator-manual/webhook.md
     - operator-manual/health.md
     - operator-manual/custom_tools.md