Merge pull request #1753 from geremyCohen/ollama_on_gke

Applied suggested modifications

Author: jasonrandrews

content/learning-paths/servers-and-cloud-computing/multiarch_ollama_on_gke/0-spin_up_gke_cluster.md

@@ -8,7 +8,7 @@ layout: learningpathall
 
 ## Project overview
 
-Arm CPUs are widely used in AI/ML use cases. In this Learning Path, you will learn how to run [Ollama](https://ollama.com/) on Arm-based CPUs in a hybrid architecture (amd64 and arm64) K8s cluster.
+Arm CPUs are widely used in Kubernetes AI/ML use cases. In this Learning Path, you learn how to run [Ollama](https://ollama.com/) on Arm-based CPUs in a hybrid architecture (amd64 and arm64) K8s cluster.
 
 To demonstrate this, you can bring up an initial Kubernetes cluster (depicted as "*1. Initial Cluster (amd64)*" in the image below) with an amd64 node running an Ollama Deployment and Service.
 
@@ -49,11 +49,7 @@ Although this will work in all regions and zones where C4 and C4a instance types
 
 8. Click on *NODE POOLS*->*Nodes*
 9. For *Series*, select *C4*
-10. For *Machine Type*, select *c4-standard-4*
-
-{{% notice Note %}}
-The chosen node types support only one pod per node. If you wish to run multiple pods per node, each node should provide about 10GB memory per pod. 
-{{% /notice %}}
+10. For *Machine Type*, select *c4-standard-8*
 
 ![Configure amd64 node type](images/configure-x86-note-type.png)
 
@@ -63,11 +59,33 @@ It will take a few moments, but when the green checkmark is showing next to the
 
 ### Connect to the cluster
 
-{{% notice Note %}}
-The following assumes you have gcloud and kubectl already installed.  If not, please follow the instructions on the first page under "Prerequisites". 
-{{% /notice %}}
+Before continuing, make sure you have *kubectl* and *gcloud* installed.  You can verify by running each command, for example, entering *gcloud* and enter:
 
-You'll first setup your newly created K8s cluster credentials using the gcloud utility.  Enter the following in your command prompt (or cloud shell), and make sure to replace "YOUR_PROJECT_ID" with the ID of your GCP project:
+```bash
+gcloud
+```
+should return
+```output
+ERROR: (gcloud) Command name argument expected.
+... 
+```
+and entering *kubectl* and enter should return:
+
+```output
+kubectl controls the Kubernetes cluster manager.
+
+ Find more information at: https://kubernetes.io/docs/reference/kubectl/
+...
+```
+If you get something similar to:
+
+```output
+command not found
+```
+
+Please follow prerequisite instructions on the first page to install the missing utilities.
+
+With prerequisites out of the way, you will next setup your newly created K8s cluster credentials using the gcloud utility.  Enter the following in your command prompt (or cloud shell), and make sure to replace "YOUR_PROJECT_ID" with the ID of your GCP project:
 
 ```bash
 export ZONE=us-central1
@@ -81,8 +99,7 @@ If you get the message:
 ```output
 CRITICAL: ACTION REQUIRED: gke-gcloud-auth-plugin, which is needed for continued use of kubectl, was not found or is not executable. Install gke-gcloud-auth-plugin for use with kubectl by following https://cloud.google.com/kubernetes-engine/docs/how-to/cluster-access-for-kubectl#install_plugin
 ```
-
-This command should help resolve it:
+This command will resolve it:
 
 ```bash
 gcloud components install gke-gcloud-auth-plugin

content/learning-paths/servers-and-cloud-computing/multiarch_ollama_on_gke/2-deploy-arm64.md

@@ -26,7 +26,7 @@ To add Arm nodes to the cluster:
 7. Select *C4A* : *c4a-standard-4* for Machine *Configuration/Type*.
 
 {{% notice Note %}}
-To compare amd64 and arm64 performance, the c4a-standard-4 is used as the arm64 equivalent of the previously deployed c4-standard-4 in the amd64 node pool.
+To compare amd64 and arm64 performance, the c4a-standard-4 is used as the arm64 equivalent of the previously deployed c4-standard-8 in the amd64 node pool.
 {{% /notice %}}
 
 ![YAML Overview](images/arm_node_config-2.png)
@@ -146,7 +146,7 @@ service/ollama-arm64-svc created
 service/ollama-multiarch-svc created
 ```
 
-2. Get the status of the pods and the services by running the following:
+2. Get the status of the nodes, pods, and services by running the following:
 
 ```bash
 kubectl get nodes,pods,svc -nollama 

content/learning-paths/servers-and-cloud-computing/multiarch_ollama_on_gke/3-perf-tests.md

@@ -72,14 +72,15 @@ The output is similar to:
 
 ```output
 ...
-1023,13],"total_duration":15341522988,"load_duration":16209080,"prompt_eval_count":32,"prompt_eval_duration":164000000,"eval_count":93,"eval_duration":15159000000}
-Tokens per second:  6.13
+"prompt_eval_duration":79000000,"eval_count":72,"eval_duration":5484000000}
+Tokens per second:  13.12
 
 Pod log output:
-[pod/ollama-arm64-deployment-678dc8556f-mj7gm/ollama-multiarch] 06:29:14
+
+[pod/ollama-amd64-deployment-cbfc4b865-k2gc4/ollama-multiarch] 2025-03-27T00:25:21
 ```
 
-You can see tokens per second rate measured at 6.13 (from the log output example, your actual value may vary a bit).
+You can see tokens per second rate measured at 13.12 (from the log output example, your actual value may vary a bit).
 
 Next, run the same inference on the arm64 node with the following command:
 
@@ -94,10 +95,13 @@ Visually, you see the output streaming out faster on arm64 than on amd64. Look a
 Tokens per second:  14.47
 
 Pod log output:
-[pod/ollama-arm64-deployment-678dc8556f-mj7gm/ollama-multiarch] 06:46:35
+
+[pod/ollama-arm64-deployment-678dc8556f-md222/ollama-multiarch] 2025-03-27T00:26:30
 ```
 
-The output shows a more than a 2X performance increase between arm64 and amd64.
+The output shows more than a 15% performance increase of arm64 over amd64.
+
+### Notes on Evaluating Price/Performance
 
 ### Price performance notes
 

content/learning-paths/servers-and-cloud-computing/multiarch_ollama_on_gke/_index.md

@@ -1,5 +1,5 @@
 ---
-title: Use GKE to run Ollama on arm64 and amd64 nodes using a multi-architecture container image
+title: Run Ollama's multi-arch container image on GKE with arm64 and amd64 nodes. 
 
 minutes_to_complete: 30