My Self Reflection about the GCP(Google Cloud Platform) Workshop conducted by LinuxWorld Informatics Private Ltd.
I just attended a two-day workshop about the Google Cloud Platform (GCP) conducted by LinuxWorld Informatics Private Limited under the mentorship of World Record Holder Mr. Vimal Daga, and I am writing this article as a self-reflection about all the cool new things I learned over these 8-10 hours, even though I was unable to practically implement them owing to a lack of the resources to create a GCP billing account.
The Physical Infrastructure of the Google Cloud Platform:
Google is a gigantic corporation and is capable of investing without hold in presumably any idea or project. As a result, the Google Cloud Platform has a wide network of physical installations all over the world, where each designated region usually has three cloud servers. These installations have digital storage capacities so unfathomable that they are often boasted to be infinite.
This worldwide distribution of infrastructure is interconnected via cables and wireless networks belonging solely to Google, thus enabling the establishment of a secure interconnected pool of resources. These resources are provided to users as on-demand services such as Compute-As-A-Service, Storage-As-A-Service and Platform-As-A-Service, all provided with Google's own security and dependability.
One of the most alluring features of GCP is that it provides the user full control over the physical locations where the instances are deployed. This ensures not only a better access to deployed services in different regions of the world, but also security against the unpredicatable nature of the physical location,which presents a risk to data stored in a location or access to the same due to natural or other factors.
Creating a GCP account:
To create a GCP account, one needs to have a credit card which, as far as I learned by experiment:
- Should be associated with either MasterCard or Visa.
- Should have the ability for recurring payments.
- Should not be a prepaid card.
If you have a card which meets these requirements and a GMail account, then the further process of creating your GCP account is extremely simple and can be done with a few clicks.
When you first create your account, GCP provides you with a free trial of three months bearing a free credit of $300.
The Google Compute Engine (GCE):
The first thing discussed during the workshop was the Google Compute Engine(GCE), which allows the user to create virtual machines,also known as instances,using an Operating System of the user's choice.
To run an instance, the user needs to pay an hourly fee depending upon the resources consumed. During your free trial, this fee is deducted from your $300 Dollar credit, and it is advisable to keep terminating instances once the required tasks are completed to preserve your trial credits and enable you to explore more.
Virtual Private Cloud (VPC):
A Virtual Private Cloud or VPC is a digital equivalent of a corporation's internal network. It allows us to create and control the networks that our instances are connected to. This enables the user to connect an instance to other instances via private connections and/or host public connections.
By default,Google Cloud provides a preset VPC for every instance upon creation. However, if the user so desires, the option to create a new personalized VPC with rules of the user's own choice is available. Similarly, the rules for the external firewall and load balancers are also changeable as per the user's requirements.
The Google Kubernetes Engine (GKE):
Kubernetes is a docker-based containerization framework developed by Google which, in addition to the ability to create containers ('pods') using Operating System images from Docker Hub, also boasts a functionality to actively monitor the deployed pods in real-time and handle any discrepancy arising due to a syste-crash in one of the pods, as well as the capability to deploy new pods upon an increase in incoming network traffic using load balancers.
The Google Cloud Platform offers Kubernetes instances as a form of Containerization-As-A-Service. As Google is the original creator of Kubernetes, it exhibits an exclusive Kubernetes-optimized Operating System for GKE instances, thus providing users with a comfortable and perfectly-enabled containerization system which basically means the essential services without redundant extra baggage.
To use the Google Kubernetes Engine,one can either use the google cloud terminal or simply install the kubectl application to be able to control it remotely. While an internal load-balancer is available, GCP provides an external load balancer for the user to customize and deploy. The user can create a node pool based on either regional or zonal distribution to deploy nodes in different physical locations.
The Google App Engine (GAE):
The largest effect of cloud-computing and any related services is almost invariably seen in the field of Developer Operations or DevOps. While the Google Kubernetes Engine provides a viable method of Continuous Delivery functionality, the Google App Engine is used to provide Continuous Integration.
To use GAE, the developers needs to install and set up the Google Cloud SDK on their systems, linking it to their GCP accounts. The GAE is deployed on Google Cloud with internal environment specifications selected by the user.
Once the set up is completed, the developer can focus on development and use the "gcloud app deploy" command to deploy the new versions of the app on the App Engine. The environment specifications are decided by the app.yaml file in the concerned directory.
Conclusion:
Thus, the workshop conducted over a period of 8-10 hours came to an end with the attendees having gained knowledge which was far deeper and more diverse than initially expected. We gained an insight to some highly productive and industrially useful services provided by the cloud infrastrcture developed by one of the world's largest and most popular tech corporations.