Use AWS accelerated AMIs (#46)

* use aws accelerated amis

* update eks accelerated ami ref link

Author: bd-g

charts/deepgram-self-hosted/CHANGELOG.md

@@ -6,6 +6,10 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 
 ## [Unreleased]
 
+### Changed
+
+- AWS samples updated to take advantage of new [EKS accelerated AMIs](https://aws.amazon.com/about-aws/whats-new/2024/10/amazon-eks-nvidia-aws-neuron-instance-types-al2023/), which bundle the required NVIDIA driver and toolkit instead of being installed by the NVIDIA GPU operator 
+
 ## [0.6.0] - 2024-09-27
 
 ### Added

charts/deepgram-self-hosted/samples/01-basic-setup-aws.cluster-config.yaml

@@ -40,7 +40,7 @@ managedNodeGroups:
     desiredCapacity: 0
     maxSize: 8
     instanceType: g6.2xlarge
-    amiFamily: Ubuntu2204
+    amiFamily: AmazonLinux2023
     labels:
       k8s.deepgram.com/node-type: engine
       k8s.amazonaws.com/accelerator: nvidia-l4
@@ -58,7 +58,7 @@ managedNodeGroups:
     desiredCapacity: 0
     maxSize: 2
     instanceType: c5n.xlarge
-    amiFamily: Ubuntu2204
+    amiFamily: AmazonLinux2023
     labels:
       k8s.deepgram.com/node-type: api
     iam:
@@ -70,7 +70,7 @@ managedNodeGroups:
     desiredCapacity: 0
     maxSize: 2
     instanceType: t3.large
-    amiFamily: Ubuntu2204
+    amiFamily: AmazonLinux2023
     labels:
       k8s.deepgram.com/node-type: license-proxy
     iam:

charts/deepgram-self-hosted/samples/01-basic-setup-aws.values.yaml

@@ -140,3 +140,15 @@ cluster-autoscaler:
   autoDiscovery:
     clusterName: "deepgram-self-hosted-cluster"
   awsRegion: "us-west-2"
+
+gpu-operator:
+  enabled: true
+
+  # If using EKS accelerated AMIs based on Amazon Linux 2023 (default with Deepgram guides as of Oct 2024),
+  # driver and toolkit come bundled with the AMI.
+  #
+  # If using other AMIs, such as Ubuntu, you should re-enable the driver and toolkit below.
+  driver:
+    enabled: false
+  toolkit:
+    enabled: false