Back to Course

GCP PlayCloud Labs

0% Complete
0/0 Steps
  1. GCP PlayCloud Labs
  2. Guided Lab: Creating a Cloud Storage Bucket
  3. Guided Lab: How to Launch a GCP Compute Engine Linux Instance
  4. Guided Lab: Creating a Cloud SQL Instance
  5. Guided Lab: Creating a Cloud NAT Gateway
  6. Guided Lab: Running SQL Commands in Cloud SQL Studio
  7. Guided Lab: Hosting a Static Website in Google Cloud Storage Bucket
  8. Guided Lab: Creating and Restoring Cloud SQL Backups
  9. Guided Lab: Vertically Scaling a VM Instance
  10. Guided Lab: Creating a Custom Image from a VM Instance with Web Server in Google Cloud
  11. Guided Lab: Setting up a Web Server on a VM Instance
  12. Guided Lab: Creating a Spot VM Instance
  13. Guided Lab: Exploring Instance Metadata in Google Cloud
  14. Guided Lab: Setting Up and Managing a Database on a VM Instance
  15. Guided Lab: Integrating Cloud SQL Database instance with a VM instance
  16. Guided Lab: Connecting Cloud SQL Database with MySQL Workbench (Local)
  17. Guided Lab: Creating a VM Using Instance Templates
  18. Guided Lab: Installing WordPress on an Ubuntu VM Instance with LEMP Stack
  19. Guided Lab: Reserving or Promoting a Static IP Address for a VM Instance
  20. Guided Lab: Creating a Google Kubernetes Engine (GKE) Cluster
  21. Guided Lab: Protecting Data on Cloud Storage Bucket Against Accidental Delete and Overwrite Using Object Versioning
  22. Guided Lab: Connecting to a Kubernetes Engine Cluster
  23. Guided Lab: Creating and Managing Instance Groups in Compute Engine
  24. Guided Lab: Configuring Shielded VM Options
  25. Guided Lab: SSH Access to GCP VM Instance from Local Machine using SSH Key Pair
  26. Guided Lab: Deploying a Simple Web Application on GKE
  27. Guided Lab: Configuring Firewall Rules to Secure and Access a VM
  28. Guided Lab: Establishing VPC Peering for Secure Cross‑Network Communication
  29. Guided Lab: Creating a Custom Virtual Private Cloud (VPC)
  30. Guided Lab: Using Startup Scripts in GCP VM Instances
  31. Guided Lab: Deploying a LAMP Stack on a Compute Engine VM
  32. Guided Lab: Managing Cloud Storage Buckets via SSH Commands
  33. Guided Lab: Guarding Your VM with Deletion Protection
  34. Guided Lab: Using Cloud Storage Lifecycle Rules to Automate Object Management
  35. Guided Lab: Guarding Your Cloud SQL Instances with Deletion Protection
  36. Guided Lab: Exploring Google Cloud Storage Classes
  37. Guided Lab: Setting Up a Linux Bastion Host on GCP
  38. Guided Lab: Uploading, Organizing, and Managing Objects in Cloud Storage
  39. Guided Lab: Creating an Application Load Balancer
  40. Guided Lab: Creating a Network Load Balancer
  41. Guided Lab: Creating VM Snapshots and Restoring a VM from a Snapshot
Lesson 12 of 41
In Progress

Guided Lab: Creating a Spot VM Instance

Description

Spot VMs in Google Cloud offer substantial cost savings—up to 91% off for many machine types, GPUs, TPUs, and Local SSDs—compared to standard VMs. Because they run on spare Compute Engine capacity, they can be preempted at any time when resources are needed elsewhere. During preemption, Compute Engine will either stop the VM (default) or delete it, depending on the termination action you set. Availability varies with overall demand, and Spot VMs have no guaranteed runtime unless you configure a time limit.
For workloads that are fault‑tolerant and can handle interruptions, Spot VMs can dramatically lower costs. A common example is batch processing: if some VMs are preempted, the job may slow down but continues running. This allows you to complete tasks without overloading existing VMs or paying full price for additional standard instances.

In this guided lab, you will learn how to create a Spot VM instance in Google Cloud and verify that the instance is running.

Prerequisites

To ensure the successful completion of this lab, you must have prior experience in creating VM instances and be familiar with their essential components. If you feel that your knowledge in this area is insufficient, we highly recommend taking this lab to gain the necessary understanding:

Objectives

In this lab, you will:

  • Create a Spot VM instance.
  • Connect to the instance via SSH.
  • Verify that the instance is running and accessible.

Lab Steps

Create a Spot VM Instance

1. In the Google Cloud Console, go to Compute Engine → VM instances → Create Instance.

2. Configure the VM:

  • Name: Enter your desired instance name.
  • Machine type: e2-micro (preset).
  • VM provisioning model: Spot (preemptible).
  • Determines how long your VMs will run. For this lab, no need to configure this.
    • Set a time limit for the VM → This option lets you define a maximum runtime (for example, 1 hour, 6 hours, or 24 hours). Once the time limit is reached, the VM will automatically be terminated. This is useful for predictable batch jobs or workloads where you know exactly how long the VM is needed. The choice is permanent once the VM is created.
    • Gracefully shut down the VM → Instead of being abruptly terminated, Google Cloud sends a preemption notice (about 30 seconds) and allows the VM to shut down cleanly. This gives your applications a chance to save state, write logs, or perform cleanup before the VM stops or is deleted.
  • Termination action: Stop or Delete (choose based on your preference). For this exercise, choose Stop.
    • Stop preserves the VM for later restart
    • Delete permanently removes it when preempted.
  • Leave other configurations at their default values.

3. Once done, click the Create button.

Connect to the Spot VM

1. From the VM instances page, click SSH next to your Spot VM.

2. Run a simple command to verify connectivity:

hostname

This should return the VM’s hostname.

Congratulations! You have successfully created a Spot VM instance in Google Cloud. You configured its availability policies, connected via SSH, and verified that the instance is running. This workflow demonstrates how Spot VMs provide a cost-effective option for running fault-tolerant workloads, with the trade-off that they can be preempted at any time. By understanding options like time limits, graceful shutdown, and termination actions, you can design workloads that balance savings with reliability.

Skip to content