Google Cloud in Action

An Overview of Google Cloud Platform

Google Cloud Platform is a “suite of cloud computing services that runs on the same infrastructure that Google uses internally for its end-user products, such as Google Search, Gmail, and YouTube.” In this article, you will find a short overview of Google Cloud Platform (GCP) and some practical usages, including using BigQuery for searching in log files and building a gaming machine in the cloud.

Getting started with Google Cloud Platform

To get started with GCP is simple: create an account on Among other registration details you will need a valid credit card. At the time this article was written Google offered $300 cloud credit that could be used for 1 year. So, you get to test the cloud capabilities for free until the credit expires, which is really nice.

Google Cloud Platform Short Overview

Google offers a variety of cloud capabilities. Please refer to the picture below for a brief summary. You can find more information in this course: Google Cloud Platform Fundamentals: Core Infrastructure.

Image from

Using BigQuery for day-to-day work

According to Google’s website, Google BigQuery is a “serverless, highly scalable, and cost-effective cloud data warehouse designed to help you make informed decisions quickly, so you can transform your business with ease”.

It’s core features:

  • Quickly analyze gigabytes to petabytes of data using ANSI SQL at blazing-fast speeds, and with zero operational overhead
  • Efficiently run analytics at scale with a 26%–34% lower three-year TCO than cloud data warehouse alternatives
  • Seamlessly democratize insights with a trusted and more secure platform that scales with your needs

By using BigQuery you can upload some log files into a DataSet and then use a SQL query like syntax in order to find information. While the log file is small, as you can see in the example, the benefit of using BigQuery increases as the log file size increases, and traditional text tools (like Notepad++) are becoming harder to use.

Creating a dataset

We can create a dataset by navigating to BigQuery menu and selecting Create Dataset option.

Creating a table and importing data

In order to store log data, we need to create a table in our dataset.

GCP can auto-detect the table schema based on the properties from our json file. The data in our test json file can be visualized here.

We can preview the data added in the table:The order of the rows from our json file was not preserved after importing data in the table.

This is a minor inconvenience, but we can order the data using SQL-like query:
SELECT * FROM `LogDataset.LogTable` ORDER BY EventTime ASC
Searching in the table
We can search for a text in the message, let’s say ‘info.Building a gaming machine in the cloud

It was exciting to learn that there are cloud gaming features provided by GCP.

Here are the steps to perform in order to build a gaming machine in the cloud. We can create a VM from the VM menu in the Compute Engine panel.The monthly estimate for the VM varies on the region selected. For example, for a n1-standard-1 machine with 1 vCPU and 3.75 GB memory the cost per month was 31.75 $ for europe-west3 (Frankfurt) region and 24.67 $ for us-central1 (Iowa).

There are a multitude of interesting options when creating a VM.

The ability to choose a GPU is quite interesting. We have the option of picking between 5 NVidia GPUs: 

NVidia Tesla K80

NVidia Tesla P4

NVidia Tesla T4

NVidia Tesla V100

NVidia Tesla P100

So, let’s go ahead and create a gaming VM, using Tesla P100, 8 CPUs, 30 GB RAM.

The GPUs are not available in all regions, for more details about availability please check:

Another option for creating the VM is Google Marketplace:

We have to make sure that the requested resources (video cards) are available. For this example, the quota was set to 0 and an error message was received related to it.

We can increase the quota limit from the quotas page. A request must be made for this.

Quotas represent limits of various cloud resources you have access to. A free trial account may need to be updated before requesting a quota increase.

Google assures us that they won’t charge us until the $300 credits expire.

After editing the quota and adding our contact details (email and phone) we can submit the request.

Google Cloud support is responsive, so after increasing the quota to 2, the machine was able to be created using the Google Marketplace. Quake 2 RTX demo was installed and it ran successfully at around 35 FPS in FullHD resolution. So there it is, gaming in the cloud!

The VM instance was deleted after the tests so that no additional billing costs are generated.

Final Words

Google Cloud provides some nice features for both developers and gamers alike. I hope this article encourages you to explore its possibilities.

Resources and Useful Links

To learn more about GCP, check out this course: Google Cloud Platform Fundamentals: Core Infrastructure

BigQuery home page:

Loading json data in BigQuery:

You can download the Quake 2 RTX demo here:,3.html

Background Image