Skip to content
This repository was archived by the owner on Mar 31, 2021. It is now read-only.

Latest commit

 

History

History
68 lines (44 loc) · 3.01 KB

gcp_setup.md

File metadata and controls

68 lines (44 loc) · 3.01 KB

Cloud environment setup

Here we detail how to install gcloud utils on your machine to interact with a project on Google Cloud Platform.

1 Create a Google Cloud Platform project

Creating a GCP project is pretty straightforward. Here, a project simply means a context in which computing resources are managed. If you are completely new to GCP here are a few screenshots to guide you through:

  1. Create a GCP project named tensorflow-serving: Select the down arrow next to the project

  2. Type the name of your project Type the name of your project

  3. Wait a few seconds for the project to be created and click on it in the notifications center: Click on your project in the notification center

Ok you are now working on the tensorflow-serving GCP project! tensorflow-serving project selected

2 Install the gcloud command line interface

gcloud CLI enable you to interact with the resources of your GCP account. Here is the installation process,

On MacOS

  1. Download the gcloud archive file https://dl.google.com/dl/cloudsdk/channels/rapid/downloads/google-cloud-sdk-231.0.0-darwin-x86_64.tar.gz

  2. Extract the archive to any location on your file system; preferably, your home directory. This can be achieved by opening the downloaded .tar.gz archive file in the preferred location.`

On Linux

  1. Curl the gcloud archive
curl -O https://dl.google.com/dl/cloudsdk/channels/rapid/downloads/google-cloud-sdk-231.0.0-linux-x86_64.tar.gz
  1. Extract the archive to any location on your file system; preferably, your Home folder. On Linux, you can extract the archive file by running this command:
tar zxvf [ARCHIVE_FILE] google-cloud-sdk

Common steps:

  1. Go to the google-cloud-sdk folder and run ./install.sh you are prompted a few questions, answer and proceed. Restart your terminal for the changes to take effect gcloud setup on macOS

  2. We now have to initialize gcloud to describe which project to work on by default. We just created a tensorflow-serving project in the previous section, let's use it! Run gcloud init: a prompt appears: you login to your GCP account and select the working project to be tensorflow-serving.

gcloud init

gcloud init choose project

  1. You may also need kubectl to manage your kubernetes cluster on GCP, install it with gcloud components install kubectl.

Great you now have your default gcloud configuration!

Resources: