From f446b3075a657be72a4f533a7c7a3d0da3147923 Mon Sep 17 00:00:00 2001
From: Taylor Dolezal <onlydole@gmail.com>
Date: Wed, 6 Mar 2024 17:24:18 -0800
Subject: [PATCH] Add instructions and scripts

Signed-off-by: Taylor Dolezal <onlydole@gmail.com>
---
 README.md                  | 64 +++++++++++++++++++++++++++++++++++++-
 kubernetes/deployment.yaml |  1 +
 shutdown.sh                | 25 +++++++++++++++
 startup.sh                 | 19 +++++++++++
 4 files changed, 108 insertions(+), 1 deletion(-)
 create mode 100755 shutdown.sh
 create mode 100755 startup.sh

diff --git a/README.md b/README.md
index fd1e5e6..722e1a1 100644
--- a/README.md
+++ b/README.md
@@ -1 +1,63 @@
-# LLMs In Action
+# LLMs in Action: A Cloud Native Story
+
+## Prerequisites
+
+- [Docker](https://docs.docker.com/install/)
+  - Docker is a platform for developers and sysadmins to develop, ship, and run applications. Docker enables you to separate your applications from your infrastructure so you can deliver software quickly.
+- [Ollama](https://ollama.com/)
+  - Ollama is a Language Model as a Service (LMaaS) that provides a RESTful API for interacting with large language models. It's a great way to get started with LLMs without having to worry about the infrastructure.
+- [kind](https://kind.sigs.k8s.io/)
+  - kind is "Kubernetes in Docker," used by the Kubernetes project to help test features and run integration tests. It turns out it's a handy way for anyone to spin up a cluster quickly. Big thank you to @bentheelder for developing it 👏🏼 👏🏼
+- [kubectl](https://kubernetes.io/docs/tasks/tools/install-kubectl/)
+  - kubectl is how you interact with k8s using the command line. This allows you to become a k8s whisperer :-)
+
+With Ollama installed on your machine, you will need to pull the LLaVa model by running
+
+```sh
+ollama pull llava
+```
+
+You can verify that the model is installed by running
+
+```sh
+ollama list
+
+NAME            ID           SIZE   MODIFIED
+llava:latest    8dd30f6b0cb1 4.7 GB 17 seconds ago 
+```
+
+## Startup
+
+We have crafted a few scripts to make this demo run as quickly as possible on your machine once you've installed the prerequisites.
+
+This script will:
+
+- Create a kind cluster
+- Apply the Kubernetes manifests we need for our demo
+- Use port-forwarding to help us access our service in the browser so we can take photos and describe them with LLaVa (Large Language and Vision Assistant)
+
+```sh
+./startup.sh
+```
+
+To access the service, open your browser and navigate to [http://localhost:8501](http://localhost:8501)
+
+## Shutdown
+
+To shut down the demo, run the following command, which will:
+
+- Remove the Kubernetes manifests
+- Remove the port-forwarding
+- Delete the kind cluster
+
+```sh
+./shutdown.sh
+```
+
+## Operating System Information
+
+This demo has been tested on the following operating systems and will work if you have the prerequisites installed.
+
+- macOS
+- Linux
+- Windows
diff --git a/kubernetes/deployment.yaml b/kubernetes/deployment.yaml
index 65351b8..ac749d3 100644
--- a/kubernetes/deployment.yaml
+++ b/kubernetes/deployment.yaml
@@ -17,6 +17,7 @@ spec:
       containers:
         - name: keynote
           image: onlydole/llm-in-action
+          imagePullPolicy: IfNotPresent
           ports:
             - containerPort: 8501
           env:
diff --git a/shutdown.sh b/shutdown.sh
new file mode 100755
index 0000000..a67e00e
--- /dev/null
+++ b/shutdown.sh
@@ -0,0 +1,25 @@
+#!/bin/bash
+
+echo "Deleting the Kubernetes manifest for this demo..."
+kubectl delete -f kubernetes/
+
+# Use pgrep to find the PIDs of the processes with 'kubectl port-forward' command
+pids=$(pgrep -f "kubectl port-forward")
+
+if [ -z "$pids" ]; then
+    echo "No kubectl port-forward processes found."
+else
+    # Display the PIDs and kill them
+    for pid in $pids; do
+        echo "Terminating process with PID: $pid"
+        kill $pid
+        if [ $? -eq 0 ]; then
+            echo "Successfully terminated process with PID: $pid"
+        else
+            echo "Failed to terminate process with PID: $pid. You may need to run the script as root or use 'sudo'."
+        fi
+    done
+fi
+
+echo "Delete the kind cluster to clean up our machine..."
+kind delete cluster -n llm
\ No newline at end of file
diff --git a/startup.sh b/startup.sh
new file mode 100755
index 0000000..e31562e
--- /dev/null
+++ b/startup.sh
@@ -0,0 +1,19 @@
+#!/bin/bash
+
+echo "Create the kind cluster..."
+kind create cluster --config cluster.yaml
+
+echo "Applying the Kubernets manifests..."
+kubectl apply -f kubernetes/deployment.yaml
+kubectl apply -f kubernetes/service.yaml
+
+# Wait for the pod starting with "keynote" to be running
+while [[ $(kubectl get pods --no-headers | grep 'keynote' | awk '{print $3}') != "Running" ]]; do
+  echo "Keynote demo is starting, make some noise..."
+  sleep 10
+done
+
+echo "Forwarding the Keynote pod..."
+kubectl port-forward svc/keynote 8501:8501 &
+
+echo "Keynote demo is running, it's time to present! Open your browser to http://localhost:8501"
\ No newline at end of file