GSoC 2020: Getting familiar with the playground

Rajitha Warusavitarana
Leopards Lab
Published in
3 min readMay 18, 2020

In the previous article, I explained how I got into open-source, why I decided to contribute to the Node Cloud project and how I got selected to Google Summer of Code. In this article, I will give a small introduction about the Node Cloud project since some of you might not know about it. Afterwards, I will give an update on what I have been doing.

The Node Cloud project

In the present developers tend to use multiple cloud platforms in a particular project for various reasons. An example scenario would be using Google Cloud AutoML for machine learning and storing the outcome of the machine learning model in the Azure Cosmos DB. In this scenario, developers have to switch between different cloud providers. This is where Node Cloud comes into play, it helps developers easily switch between cloud providers. Currently, Node cloud supports AWS, Google Cloud, Azure, Digital Ocean and Alibaba Cloud. In this summer my aim is to improve Node Cloud by implementing new services in AWS, Google Cloud and Azure plugins.

Getting familiar with the playground

It’s always better to get familiar with the playground before starting a game on it. Since the Node Cloud project is an existing project, the community bonding period is very useful for me to get familiar with the current codebase. Because I have been contributing to this project since 2019, I was already familiar with the Node Cloud AWS plugin and architecture of the system. Therefore I decided to work in another plugin of NodeCloud which is the Google Cloud plugin and integrate a new service to it. In addition to that, I have created Github issues in the relevant repositories regarding the tasks that I will be working on this summer.

Integrating Cloud AutoML in the Node Cloud GCP plugin

Cloud AutoML is a cloud service provided by Google Cloud platform which helps to train high-quality machine learning models specific to your business needs. The awesome part here is, you are not required to be an expert in machine learning to use this service. I was able to integrate this service in the Node Cloud GCP plugin and test in Google Cloud with my free trial account. While I was working on this I got into some blockers as well. As usual, I searched for an answer in Stack Overflow and found an unanswered question related to the same issue but in a different stack. Looking more into the problem I got sorted the issue and was able to answer the question in Stack Overflow as well 😀. Currently, the work I have done is in the review stage. I will let you know about the status regarding this service in my next article. So, that’s pretty much it of what I have done so far. See you again in the next article :)

Good Bye…

--

--