Skip to main content

Introduction to Google Compute Engine

What is a Virtual Machine?​

A Virtual Machine (VM) is a software-based emulation of a physical computer that runs an operating system and applications. Think of it as a computer within a computer, where:

  • It has its own CPU, memory, storage, and network interface
  • It runs its own operating system independently
  • It can run applications just like a physical computer
  • It's isolated from other VMs on the same physical machine

Key Characteristics of VMs:​

  • Isolation: Each VM operates independently
  • Portability: Can be moved between physical machines
  • Scalability: Easy to create, modify, or delete
  • Resource Sharing: Multiple VMs can run on a single physical machine

Alternatives to Virtual Machines​

  1. Physical Servers

    • Traditional dedicated hardware
    • Direct access to physical resources
    • No virtualization overhead

  2. Containers

    • Lightweight, portable units of software
    • Share the host OS kernel
    • Faster startup and lower overhead

  3. Serverless Computing

    • Event-driven execution
    • No server management
    • Pay-per-use pricing

What is Google Compute Engine?​

Google Compute Engine (GCE) is Google Cloud's Infrastructure as a Service (IaaS) offering that lets you create and run virtual machines on Google's infrastructure.

Basic Features:​

  1. Machine Types

    • Predefined configurations (e.g., n1-standard-1)
    • Custom machine types
    • GPU-enabled instances
    • Memory-optimized instances

  2. Storage Options

    • Persistent Disks (SSD and HDD)
    • Local SSDs
    • Cloud Storage integration

  3. Networking

    • Virtual Private Cloud (VPC)
    • Load balancing
    • Cloud CDN
    • Cloud NAT

  4. Management Tools

    • Google Cloud Console
    • Command-line interface (gcloud)
    • REST APIs
    • Infrastructure as Code (Terraform)

How to Manage Google Compute Engine​

Google Cloud Console provides a web-based interface to create, manage, and monitor virtual machines without using the command line. Follow these steps to manage GCE through the console.

1. Creating a New VM Instance​

To create a virtual machine instance using the Google Cloud Console:

  1. Navigate to Compute Engine

  2. Click on "Create Instance"

    • Click the "Create Instance" button at the top.

  3. Configure Your VM Instance

    • Name: Enter a unique name for your instance.
    • Region & Zone: Choose the region (e.g., us-central1) and zone (us-central1-a) based on your latency and compliance needs.
    • Machine Type:
      • For general workloads: e2-standard-2
      • For compute-intensive tasks: c2-standard-4
      • For memory-intensive tasks: m2-ultramem-160
    • Boot Disk:
      • Click "Change" to select an OS (Ubuntu, Debian, CentOS, Windows, etc.).
      • Choose disk type (SSD for high performance, Standard HDD for lower cost).
    • Firewall Settings:
      • Check "Allow HTTP traffic" and "Allow HTTPS traffic" if serving a web application.
    • Click "Create" to launch the VM.

  4. Access the VM

    • Once created, find your VM under Compute Engine > VM Instances.
    • Click "SSH" next to your instance name to open an SSH session directly in the browser.

2. Managing VM Resources​

Starting & Stopping Instances​

  • In the VM Instances list, locate your instance.
  • Click the three-dot menu (⋮) next to your instance.
  • Choose:
    • Start – If the instance is stopped.
    • Stop – To shut down the instance.
    • Restart – To reboot the instance.

Resizing a Machine Type​

  • Click on your instance name.
  • Under Machine Configuration, click Edit.
  • Select a new machine type (more vCPUs, RAM, or GPU-enabled options).
  • Click Save and confirm the restart.

Attaching Additional Storage​

  • Navigate to Compute Engine > Disks.
  • Click "Create Disk".
  • Choose:
    • Type: Standard persistent disk, SSD, or local SSD.
    • Size: Define the disk capacity.
  • Click Create.
  • To attach to a VM:
    • Go to VM Instances, select your instance, click Edit.
    • Under Additional Disks, click Attach Existing Disk.
    • Select the disk and click Save.

3. Configuring Networking​

Setting Up a Static External IP​

By default, VMs have an ephemeral IP that changes when restarted. To assign a static IP:

  • Navigate to VPC network > External IP addresses.
  • Click "Reserve Static Address".
  • Assign it to your VM and click Reserve.

Creating a Firewall Rule​

  • Go to VPC Network > Firewall.
  • Click "Create Firewall Rule".
  • Define:
    • Name: e.g., allow-web-traffic
    • Targets: Select All instances or specify your instance.
    • Source IP Ranges: 0.0.0.0/0 (for public access) or restrict to a specific IP range.
    • Protocols & Ports: Choose TCP and enter ports like 80, 443 for web servers.
  • Click Create.

4. Monitoring & Logs​

Viewing Metrics & Performance​

  • Navigate to Compute Engine > VM Instances.
  • Click on an instance to see CPU usage, memory, disk, and network metrics.
  • Click Monitoring to set up alerts for CPU usage, disk IO, and uptime.

Viewing Logs​

  • Go to Logging > Logs Explorer.
  • Filter logs by:
    • Resource type: GCE VM Instance
    • Log severity: Errors, warnings, info, etc.
  • Use Cloud Monitoring for real-time logs.

5. Security Best Practices​

Enabling IAM & Access Controls​

  • Go to IAM & Admin > IAM.
  • Assign roles:
    • Compute Admin – Full access to GCE.
    • Compute Viewer – Read-only access.
    • Custom Roles – Restrict access to specific actions.

Enabling Shielded VMs (For Security)​

  • While creating a VM, enable Shielded VM Features for protection against rootkits and firmware attacks.

Setting Up SSH Key Authentication​

  • Go to Compute Engine > Metadata.
  • Click SSH Keys > Add Key.
  • Paste your public SSH key (~/.ssh/id_rsa.pub).
  • Click Save.

6. Backups & Snapshots​

Creating a Snapshot (Backup)​

  • Navigate to Compute Engine > Snapshots.
  • Click "Create Snapshot".
  • Select the disk to back up.
  • Define the snapshot schedule (e.g., daily, weekly).
  • Click Create.

Restoring a VM from a Snapshot​

  • Create a new disk from a snapshot under Compute Engine > Disks.
  • Attach it to a new or existing VM.

7. Setting Up a Load Balancer (For Multiple VMs)​

If running a scalable app, use a Load Balancer to distribute traffic:

  • Go to Network Services > Load Balancing.
  • Click Create Load Balancer.
  • Select HTTP(S) Load Balancer.
  • Configure a Backend Service with VM instances.
  • Set up a Frontend with a public IP.
  • Click Create.

Additional Technical Information​

Instance Groups​

  • Managed Instance Groups (MIGs): Automatically scales VMs.
  • Unmanaged Groups: Manually controlled VMs.

Preemptible vs. Spot Instances​

  • Preemptible Instances: Lower-cost but may be terminated by Google.
  • Spot Instances: Replaces preemptibles with a bidding system.

Custom Machine Types​

  • Create VMs with exact CPU & RAM needs instead of predefined sizes.

GPU & AI Compute​

  • Attach NVIDIA GPUs for ML workloads.
  • Use TPUs for AI-based tasks.

Hybrid & Multi-Cloud​

  • Connect GCE to on-prem or other clouds using Anthos.

I have a created a VM Instance for Google Compute Engine. How can I SSH into it?​

1. Using Google Cloud Console (Browser SSH)​

This is the easiest way and requires no setup.

  • Go to the Google Cloud Console.
  • Navigate to Compute Engine > VM Instances.
  • Find your VM instance in the list.
  • Click the SSH button next to your instance.
  • A new browser window opens with a terminal session connected to your VM.

2. Add Your Public Key to Google Compute Engine​

  • Go to Google Cloud Console.
  • Navigate to Compute Engine > VM Instances.
  • Click on your VM instance.
  • Click Edit.
  • Scroll down to the SSH Keys section.
  • Click Add Item.
  • Copy and paste your public SSH key from your local machine 
       cat ~/.ssh/id_rsa.pub
  • Click Save.

Conclusion​

Google Compute Engine provides a powerful and flexible platform for running virtual machines in the cloud. Whether you're running a small website or a large-scale enterprise application, GCE offers the tools and features needed to build, deploy, and manage your workloads effectively.

Remember to:

  • Start with the right machine type
  • Implement proper security measures
  • Monitor costs and performance
  • Follow best practices for reliability
  • Use automation tools for efficiency