From 69da6d0fc8588341bc321f3725849ec17b16b536 Mon Sep 17 00:00:00 2001
From: Kristian Hartikainen <hartikainen@berkeley.edu>
Date: Sun, 28 Apr 2019 14:58:51 -0700
Subject: [PATCH] [autoscaler] Remove unnecessary apt installations in docker
 commands (#4577)

* Remove unnecessary apt installations in docker commands

* Add example for different head/worker image in gcp gpu example

* Update gcp gpu example docker image to tf 1.13

* Change the VM sourceImage for gcp/example-full.yaml

* Change the gcp gpu docker VM images for consistency

* Change gcp default project id to be consistent with other examples
---
 python/ray/autoscaler/docker.py                  |  6 +-----
 python/ray/autoscaler/gcp/example-full.yaml      |  4 ++--
 .../ray/autoscaler/gcp/example-gpu-docker.yaml   | 16 ++++++++++++----
 3 files changed, 15 insertions(+), 11 deletions(-)

diff --git a/python/ray/autoscaler/docker.py b/python/ray/autoscaler/docker.py
index 7f91a8096..c1b384dee 100644
--- a/python/ray/autoscaler/docker.py
+++ b/python/ray/autoscaler/docker.py
@@ -107,11 +107,7 @@ def docker_start_cmds(user, image, mount, cname, user_options):
         image, "bash"
     ]
     cmds.append(" ".join(docker_check + docker_run))
-    docker_update = [
-        " && ".join(("apt-get -y update", "apt-get -y upgrade",
-                     "apt-get install -y git wget psmisc"))
-    ]
-    cmds.extend(with_docker_exec(docker_update, container_name=cname))
+
     return cmds
 
 
diff --git a/python/ray/autoscaler/gcp/example-full.yaml b/python/ray/autoscaler/gcp/example-full.yaml
index 405185687..957569115 100644
--- a/python/ray/autoscaler/gcp/example-full.yaml
+++ b/python/ray/autoscaler/gcp/example-full.yaml
@@ -67,7 +67,7 @@ head_node:
         initializeParams:
           diskSizeGb: 50
           # See https://cloud.google.com/compute/docs/images for more images
-          sourceImage: projects/ml-images/global/images/family/tf-1-13-gpu
+          sourceImage: projects/deeplearning-platform-release/global/images/family/tf-1-13-cpu
 
     # Additional options can be found in in the compute docs at
     # https://cloud.google.com/compute/docs/reference/rest/v1/instances/insert
@@ -90,7 +90,7 @@ worker_nodes:
         initializeParams:
           diskSizeGb: 50
           # See https://cloud.google.com/compute/docs/images for more images
-          sourceImage: projects/ml-images/global/images/family/tf-1-13-gpu
+          sourceImage: projects/deeplearning-platform-release/global/images/family/tf-1-13-cpu
     # Run workers on preemtible instance by default.
     # Comment this out to use on-demand.
     scheduling:
diff --git a/python/ray/autoscaler/gcp/example-gpu-docker.yaml b/python/ray/autoscaler/gcp/example-gpu-docker.yaml
index fda967c56..43b9d867b 100644
--- a/python/ray/autoscaler/gcp/example-gpu-docker.yaml
+++ b/python/ray/autoscaler/gcp/example-gpu-docker.yaml
@@ -23,11 +23,19 @@ autoscaling_mode: default
 # and opens all the necessary ports to support the Ray cluster.
 # Empty string means disabled.
 docker:
-    image: "tensorflow/tensorflow:1.12.0-gpu-py3"
+    image: "tensorflow/tensorflow:1.13.1-gpu-py3"
     container_name: "ray-nvidia-docker-test" # e.g. ray_docker
     run_options:
       - --runtime=nvidia
 
+    # # Example of running a GPU head with CPU workers
+    # head_image: "tensorflow/tensorflow:1.13.1-gpu-py3"
+    # head_run_options:
+    #     - --runtime=nvidia
+
+    # worker_image: "ubuntu:18.04"
+    # worker_run_options: []
+
 
 # The autoscaler will scale up the cluster to this target fraction of resource
 # usage. For example, if a cluster of 10 nodes is 100% busy and
@@ -44,7 +52,7 @@ provider:
     type: gcp
     region: us-west1
     availability_zone: us-west1-b
-    project_id: <project_id> # Globally unique project id
+    project_id: null # Globally unique project id
 
 # How Ray will authenticate with newly launched nodes.
 auth:
@@ -68,7 +76,7 @@ head_node:
         initializeParams:
           diskSizeGb: 50
           # See https://cloud.google.com/compute/docs/images for more images
-          sourceImage: projects/deeplearning-platform-release/global/images/family/tf-latest-gpu
+          sourceImage: projects/deeplearning-platform-release/global/images/family/tf-1-13-cu100
     guestAccelerators:
       - acceleratorType: projects/<project_id>/zones/us-west1-b/acceleratorTypes/nvidia-tesla-k80
         acceleratorCount: 1
@@ -91,7 +99,7 @@ worker_nodes:
         initializeParams:
           diskSizeGb: 50
           # See https://cloud.google.com/compute/docs/images for more images
-          sourceImage: projects/deeplearning-platform-release/global/images/family/tf-latest-gpu
+          sourceImage: projects/deeplearning-platform-release/global/images/family/tf-1-13-cu100
     guestAccelerators:
       - acceleratorType: projects/<project_id>/zones/us-west1-b/acceleratorTypes/nvidia-tesla-k80
         acceleratorCount: 1