adding experiment removing dns

Signed-off-by: vsoch <vsoch@users.noreply.github.com>

google/service-timing/README.md

@@ -5,3 +5,6 @@ of experiments are attempting to investigate different aspects.
 
  - [run1](run1) was my original testing done on GKE to report the issue (May 4, 2023)
  - [run2](run2) was a secondary test to run nslookup across different cluster flags (May 16, 2023)
+ - [run3](run3) ran telnet in the worker pod to look at connection times/patterns to the broker leader (index 0)
+ - [run4](run4) used a test deployment of the operator that wrapped flux start with strace
+ - [run5](run5) attempts to remove DNS by getting pod ip addresses and writing them into `/etc/hosts`

google/service-timing/run5/README.md

@@ -0,0 +1,109 @@
+# Service Timing
+
+For this experiment we want to try a test branch of the operator (hash / digest in the
+flux-operator.yaml) that eliminates DNS by writing the hostnames into `/etc/hosts`
+You can see the original runs in [run1](../run1)
+
+### Setup
+
+```console
+gcloud container clusters create flux-operator \
+   --region=us-central1-a --project $GOOGLE_PROJECT \
+   --machine-type n1-standard-2 --num-nodes=4 \
+   --tags=flux-cluster  --enable-dataplane-v2
+
+$ kubectl create namespace flux-operator
+```
+
+### Hostnames on Current Networking
+
+First let's apply the current design (with wrap added) so we can save pods (and see hostnames)
+
+```
+$ kubectl create namespace flux-operator
+$ kubectl apply -f current/flux-operator-dev.yaml
+$ kubectl apply -f current/minicluster.yaml
+```
+
+Wait for the pods to be running, then save logs and pods:
+
+```
+$ kubectl logs -n flux-operator flux-sample-0-bfmll -f > current/flux-sample-0.log
+# do for each of 1,2,3 too
+$ kubectl get pods -n flux-operator -o json > current/pods.json
+$ kubectl get nodes -o json > current/nodes.json
+```
+
+And inspect hostnames:
+
+```
+$ kubectl get pods -n flux-operator -o yaml | grep hostname
+    hostname: flux-sample-0
+    hostname: flux-sample-1
+    hostname: flux-sample-2
+    hostname: flux-sample-3
+```
+
+And run the programmatic portion:
+
+```bash
+python time-minicluster.py current
+```
+
+Then you should be able to delete the operator (and the MiniCluster goes with it).
+
+```bash
+$ kubectl delete -f current/flux-operator-dev.yaml
+```
+
+### No DNS
+
+The new design can [be seen here]() - this isn't intended to be merged, but just
+for this experiment. As an example of the config map it generates (that will be used
+to update the `/etc/hosts`:
+
+```
+$ kubectl apply -f no-dns/flux-operator-dev.yaml
+$ kubectl apply -f no-dns/minicluster.yaml
+```
+
+Wait for the pods to be running, then save logs and pods:
+
+```
+$ kubectl logs -n flux-operator flux-sample-0-bfmll -f > ./no-dns/flux-sample-0.log
+# do for each of 1,2,3 too
+$ kubectl get pods -n flux-operator -o json > no-dns/pods.json
+```
+
+```
+$ kubectl get pods -n flux-operator -o yaml | grep hostname
+    hostname: flux-sample-0
+    hostname: flux-sample-1
+    hostname: flux-sample-2
+    hostname: flux-sample-3
+```
+
+And delete
+
+```bash
+$ kubectl delete -f ./no-dns/minicluster.yaml
+```
+
+Run programatically:
+
+```bash
+python time-minicluster.py no-dns
+```
+
+Don't forget to clean up!
+
+```bash
+$ gcloud container clusters delete flux-operator
+```
+
+#### Compare Flux Start Times
+
+Now we want to compare the same above, but programatically across runs.
+
+
+