Not many years ago around us was full of CDs, DVDs, Flash Drives and Hard Drives and it was hard to imagine life without them. But now it seems very hard to see any of those things around us, what happened then? The answer is Hard Drives moved to clouds! For free or for a small fee we had access to numerous storage somewhere outside our computers that pushed us to upload every thing that we used to download.
But this is not end of the story and HDDs are not the last part of our computers that could fly! Now, CPUs start to move to clouds and this is beginning of the cloud computing journey. Following are some definitions of cloud computing:
“Cloud computing is the delivery of computing services—including servers, storage, databases, networking, software, analytics, and intelligence—over the Internet (“the cloud”) to offer faster innovation, flexible resources, and economies of scale" (from Azure website).
“Cloud computing is the on-demand delivery of IT resources over the Internet with pay-as-you-go pricing. Instead of buying, owning, and maintaining physical data centers and servers, you can access technology services, such as computing power, storage, and databases, on an as-needed basis from a cloud provider” (from AWS website).
In this tutorial we will learn how to create and access to a compute engine on Google Cloud and build a Virtual Machine (VM) that could be accessible from anywhere.
To begin, we need a Google Cloud account. Note that Google Cloud basically is not a free service but you can access some services with no cost. Follow the instructions in here to create a new compute engine (VM instance), for example f1-micro (1 vCPU, 0.6 GB memory, 10GB size) can be used with no cost. There are several OS options, but in tutorial we will use CentOS.
After creating the virtual machine (VM), click on the instance name and press edit button on the top of the page and add your SSH public key to the instance. After adding the key, you will see the username in the first column. If you don’t have a public key, open a Linux terminal in your computer and run:
Make sure to create a strong passphrase and remember it for the latter. To see the public key, use the following in the terminal prompt:
Copy the outputs and add it to the instance. Now to connect the VM instance, we only need to copy the external IP of the instance and run the following in a Linux terminal prompt and enter the public key passphrase (find more details here):
After connecting to the instance, we can change the ssh configuration in a way that connecting to the VM without a public key. To connect to the instance with a root password, open the /etc/ssh/sshd_config config file with a text editor (e.g. sudo emacs /etc/ssh/sshd_config) and comment #PasswordAuthentication no and uncomment PasswordAuthentication yes. After save and exit, run the following to apply changes:
Note: in a Debian based Linux, use sudo service ssh restart.
After these changes, we can connect to the instance by both public key passphrase (if a public key exist) or root password of the instance (if there is no public key). Note that the VM’s in cloud engine do not come with a root password setup by default, use the following to set it up:
To find the user name run whoami command (that should be same as the username pair to the public key).
Also, we can use Chrome Secure Shell App by entering the username and external IP address of the instance such that:
We can use root password to connect (because of changes in the SSH config file) but if you want to use the public key, select the related private SSH key in the Identity field. If necessary, click Import to select a private key file from your local workstation (~/.ssh/id_rsa) - if using macOS, press shift+command+. to see hidden files.
Here we go! now we can connect from a terminal prompt of a local computer that has the public key, or from a Chrome browser from any computer. Note that if multiple users are going to connect to the instance, it will be better to add a OS login to the instance by adding a metadata entry where the key is enable-oslogin and the value is TRUE.
Connecting to the instance, connect us to a computer with CentOS operating system. In the following we will learn how to update CentOS and install required software.
If you have some experiences with Linux, probably you are familiar with Debian package manger (apt). CentOS uses a different package manager called yum which is pretty similar to apt. We can check for the update by:
And update OS by:
Note: we can find OS version by cat /etc/redhat-release.
To install, update or remove packages use:
Let’s install wget, git and emacs by:
sudo yum install wget # to download files from the Internet
sudo yum install git # to access Git
sudo yum install emacs # to have Emacs text editorTo install Miniconda3 run:
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.shLets refresh the .bashrc file to add conda command to the current environment by:
If you’d prefer that conda’s base environment not be activated on startup, run:
And use the following to activate conda base later:
See here for more details about installing Miniconda and learn more about using conda in here. Also, review a tutorial about conda in here.
Note: (optional) in CentOS python command refers to Python 2. To access python3 command, we can add alias python3=~/miniconda3/bin/python3 to ~\.bashrc file. Use source ~/.bashrc to apply changes.
To install R, we can use conda to create an R environment:
To use R, first activate the r_env by conda activate ./r_env (when you are in the same directory as r_env) and run R.
Now we can download and install other required software. yum and conda package managers can be used to install different software and packages same as what we did in above to install Emacs and R. Also, we can directly install software by downloading source files and install them such as installing Miniconda in above. After installation, we can reboot the engine by:
Now I think your VM is ready, enjoy it!