Google Cloud Interview Questions

Google Cloud Platform is a set of computing services that provides infrastructure as service, platform as service, and serverless computing. Services such as cloud data storage, data analytics, compute engine and machine learning are some examples of it. First released in 2008, GCP is one of the popular cloud service platforms available in the market along with Amazon Web Service, and Microsoft Azure.

Virtual Private Cloud (VPC) in GCP is a virtual network that provides connectivity to your VM instances of compute engine, GKE (Google Kubernetes Engine) clusters, and many other resources. The VPC provides much flexibility in controlling how the workloads connect regionally or globally. A single VPC can span multiple regions without communicating across the public internet.

Some of the major features of GCP are,

• GCP provides an easy way to rightsize your Virtual machine resources such as CPU, RAM, and storage. The VM rightsizing recommendation feature gives you a glance at whether your machines are with the right size of resources.
• The Google cloud shell present with GCP has many pre-installed tools and lets you control the various processes from the shell. Some preinstalled tools are Docker, Gradle, Make, npm, nvm, pip, and more.
• You can easily create your custom machine type with varying resources of CPU, memory, storage with GCP.
• It has preemptible VMs so fault-tolerant jobs and batch jobs cost up to 70 % less than normal.
• The Cloud SQL feature present in GCP automatically checks the storage available in the database every 30 seconds and adds when it is needed.
• You can resize persistent disk in-pace without any downtime.

Some of the popular alternatives to GCP are,

• Amazon Web Services
• Microsoft Azure
• Kahu
• Platform9
• Packet
• OpenStack

Four types of development models in Cloud computing are,

• Public cloud - This type of cloud can be accessed by all on a subscription basis. The public can access resources such as OS, CPU, memory, and storage.
• Private cloud - This type of infrastructure is used by a single organization and not by the general public. They are typically more expensive to build than public clouds.
• Hybrid cloud - This infrastructure uses a combination of public and private clouds. It is used by many organizations to rapidly scale up their resources when needed.
• Community cloud - Here, multiple organizations share their resource and create a pool which is restricted to the members of the community.

Some major components of GCP are,

• Compute engine such as App Engine, Kubernetes, etc.
• Storage and Database components.
• Networking components such as VPC, Cloud Armor, and more.
• Big Data components.
• Identity and Security tools like Cloud IAM, Cloud Identity, and more.
• Management tools like Trace, logging, and debugger.
• Developer tools like Cloud test lab, container builder, and more.
• Productivity and professional tools.

Google Cloud SDK is a set of command-line tools. It is used for the development of the Google cloud. With these tools, you can access the compute engine, cloud storage, bigquery, and other services directly from the command line. It also comes with client libraries and API libraries. With these tools and libraries, you can work with VM instances, manage computer engine networks, firewalls & storage, and more.

A Paas (Platform as a Service) product, Google App Engine provides scalable services to web app developers and enterprises. With this, developers can build and deploy a fully managed platform and scale when needed.

It has support for popular programming languages such as Java, PHP, Python, C#, .Net, Go, and Node.js. It is also flexible so you can develop powerful applications.

GCP storage is a cloud storage service offered by Google. It allows you to access your data from anywhere in the world at any time. This storage is highly durable, secured, and scalable. With this storage service, you can store your personal data, application data, client’s data, and more.

Buckets are the basic containers that are used to store the data. With buckets, you can organize data, and give control access. The bucket has a globally-unique name with a geographic location where the contents are stores. It also has a default storage class that is applied to the objects which don’t have a storage class specified and are added to the bucket. There is also no limit on the creation or deletion of the buckets.

In GCP Object versioning is used to retrieve objects which are overwritten or deleted. Object versioning increases storage costs but it provides security for objects when they are deleted or overwritten. On enabling the object versioning in the GCP bucket, a noncurrent version of the object is created every time when the object is overwritten or deleted. The properties used to identify a version of the object are generation and meta generation. Generation identifies the content generation while meta generation identifies the metadata generation.

In serverless computing, the cloud service provider has a server running in the cloud and dynamically manages the allocation of resources. The provider provides the necessary infrastructure for the user to work on without any worries about the hardware. The users need to pay for the resources they have used. It simplifies the process of code deployment while removing any worries regarding scalability, maintenance for the users. It is a form of utility computing.

Load Balancing is the process of distributing the computing resources and workloads in a cloud computing environment to manage the demands. It helps to achieve high performance for lower costs as the workload demands are efficiently managed with the allocation of resources. It uses the concept of scalability and agility to improve the availability of resources to the demands. It is also used to provide health check-ups for the cloud application. This feature is offered by all the major cloud vendors like AWS, GCP, Azure, etc.

BigQuery is an enterprise warehouse service offered in the Google Cloud Platform. This highly-scalable cost-effective product has an in-memory data analysis engine and machine learning built-in. It gives you the option to quickly analyze the data in real-time and create analytical reports with the help of a data analytics engine. BigQuery can also process external data sources in object storage, transactional database, or spreadsheets.

Firebase or Google cloud messaging is a cross-platform (Android, iOS, Web) notification solution which lets you deliver and receive messages and notifications free of cost. With this, you can send messages or notify client apps or send messages to drive user reengagement. It gives you the option for versatile messages targeting such as distributing the message to single devices, to a group of devices or to subscribed devices.