Most data analysts or data scientists just want easy access to data, analytics and reporting tools. We want to be able to get on with our data insights and not worry about software or server setup and installation. The development of cloud platforms for data analytics has helped to address this. In this post, I will share my experience with Google Cloud Platform (GCP). I’m currently using it for some of the analytics examples found in solvewithdata’s knowledge base and will discuss the benefits of using Google Cloud Platform for data analytics.
What are the benefits of using Google Cloud Platform for data analytics?
GCP provides a one-stop location of data science tools.
As long as you are connected to the Internet, you can access them via your web browser from any location. The platform provides files cloud storage, easy to use database tools for efficient data storage management and access to its machine learning algorithms.
Low cost and hassle-free data analytics environment setup
It is low cost and you don’t have to worry about investing in computer hardware or software. GCP currently offers a free tier plan which is sufficient for those who don’t require much computing power or data storage. There is also a A$300 free credit for 12 months for those who need to scale up use on any GCP product. One of the things that attracted me to GCP was the ease of set up. All I needed to do was sign up and I could access enterprise-ready analytics tools through my web browser. I didn’t have to get involved with any installation or setup configuration.
What analytics tools does Google Cloud Platform offer?
Here are some useful data analytics tools offered by GCP that I found to be really helpful with my data analytics projects and workflow.
Store and access data files with free cloud storage
With the free tier plan, you get 5 GB storage with Google Cloud Storage. Store your flat file, text file data dumps here. You can then use the data warehouse tools offered by GCP to read and import these files.
Simple to setup and enterprise scalable database/data warehouse
Need a database for your analytics project? Google Cloud Platform an enterprise data warehouse on the cloud with their Google BigQuery product. All you need is a web browser and you can start creating database tables, storing your data records and performing various database table queries. I found this product to be extremely useful as I didn’t have to go through the hassle of finding a computer to install and host my database. I could query data through an easy to use web interface. In addition, Google BigQuery offers an API that allows you to select and query your data from external applications such as Python Notebooks.
The free tier plan offers 1TB queries per month and 10GB storage.
Easily call Google’s machine learning modules
Gain access to Google’s Machine Learning modules such as Google Cloud Natural Language. With this module, you will be able to analyse, predict sentiment levels or classify content from blocks of text. All your analytics tool has to do is to pass the block of text to the module and Google’s Machine Learning module will perform the necessary algorithms to produce the outputs.
Access your data on Google Cloud Platform to create custom reports and dashboards
Some analytics projects require reports and dashboards to be developed to track and monitor progress and performance. Google also provides dashboard functionality via its product Google Data Studio.
Google Data Studio integrates with the Google Cloud Platform. Data Studio allows you to easily access and visualise your data stored on Google BigQuery. In my view, this functionality adds to the benefits of using Google Cloud Platform for data analytics.
In my opinion, Google Cloud Platform provides an easily accessible foundation platform to run analytics projects. The only shortcoming I experienced was the analytics front end, particularly if you are using Python.
Jupyter Notebooks are a popular and useful data analytics front-end for the Python platform. Analysts code in Python and view the results of their analytics projects on Jupyter Notebooks. Google Cloud Platform offers access to Jupyter Notebooks. However, I’ve found it not very stable. It is also not easily accessible as you have to run some scripts to launch the notebook.
Fortunately, there are Jupyter Notebook solutions in the market. Microsoft’s Azure Machine Learning Studio offers a free solution that is easy to use and cloud-based.
Using Azure ML’s Jupyter solution and Google Cloud Platform’s APIs, you can extract and analyse your data hosted on GCP.
I will blog about more about the Microsoft’s Azure Machine Learning Studio platform in future posts and also share some examples of how to integrate the 2 platforms on solvewithdata’s knowledge base.