Cloud Computing Interview Questions - Part 2
By Kamal Jacob
Traditionally, any IT ecosystem involved significant investment in terms of cost, time and skills for its smooth running. Cloud computing helps you significantly reduce the total cost of the owner and increase the time to market. As a result, it opens up new doors of opportunity for professionals like you. To help you ace your shot at the cloud computing interviews, we have covered some very popular questions asked during cloud computing interview.
In the 2nd part of the series here are few commonly asked questions, which you should go through:
How do you guarantee a reliable/secure transaction in the cloud?
The successful cloud transaction can be measured by ensuring that the ACID properties (Atomicity, Consistency, Isolation, Durability) are met.
How can a company benefit from cloud computing?
For answering this question, it is important to understand the advantages of cloud computing over traditional computing. Some of which are as follows:
a) Pay-as-you go model helps businesses save money!
b) Cloud computing offers more secure data backup and data storage.
c) Cloud also offers the ability to host your applications and data at a geographic region of your choice.
d) Allows you to gain access to powerful server capabilities without making any significant hardware investment.
e) Platform and OS agnostic.
f) It allows you to use Software as a service (SaaS) instead of buying the licenses and then, installing them on your machines.
g) It adds sandboxing and virtualization capabilities to your workloads.
h) With the readily available software, tools, and operating systems, the overall productivity of developers increases multiple times.
i) More cost-effective as compared to the traditional on-premise model
j) Ability to scale automatically and add resources on the fly without causing outages to the system.
What is your opinion on the cost of cloud computing versus traditional on-premise computing?
In the traditional on-premise model, deployment of a workload requires a significant up-front payment in establishing the infrastructure (servers, storage, network, electricity, cooling equipment, etc.) followed by a consistently spent cost for maintenance and upgrades. In addition, you need to spend time, efforts and cost in order to ensure disaster recovery and BCP (Business Continuity Planning). In addition, challenges related to under-utilization of your resources and inventory management arise which further add up to the cost of owning the IT.
In the case of cloud computing, you can deploy a workload without owning the infrastructure at all. Instead, you can just rent the resources (such as servers, storage, network, platforms or even the software services) and leaving the cost and efforts of their maintenance, cooling, etc to the cloud service provider. In addition, because you do not own the infrastructure, you pay only for the resources you utilize. Other costs, such as disaster recovery, backups, and BCP, are an implicit part of the offerings from most renowned cloud service providers and are offered to you at a nominal cost. In addition, cloud computing eliminates the need and thus, the cost incurred in administration and inventory management of the IT resources.
Overall, it’s safe to say that cloud computing is much more cost-effective than the traditional on-premise computing model.
What are the few most popular cloud-based databases?
Among the modern databases of today, there are multiple types of databases available, such as:
1. RDBMS (Relational Database Management System)
This is the most traditional type of databases. In the cloud, there are several of them offered by various service providers. For example, RDS by AWS (Amazon Web Services) provides, MySQL, Redshift, PostgreSQL, Oracle, etc. CloudSQL by GCP (Google Cloud Platform), Azure SQL DB, Azure DB for PostgreSQL, Azure DB for MySQL are other popular examples of cloud-based RDBMS.
2. OLAP databases (Online Analytical Processing)
Azure Analysis service is an example of the OLAP services on the cloud. However, because OLAP databases are no longer very popular because of the advent of highly performant NoSQL databases, not many cloud providers offer an OLAP service.
3. Graph databases
Amazon Neptune and Neo4J are two popular choices for a cloud-based graph database. Neo4J, being one of the most popular graph DB, is available from other cloud providers like Google and Microsoft Azure as well.
4. Document database
Azure Cosmos DB, AWS DynamoDB, Google BigQuery are few examples of document databases. In addition, you could choose MongoDB as a service on AWS and Azure too.
5. Columnar database
Cassandra is a popular columnar DB. On the cloud, you could use Amazon Redshift for storing data in the columnar format.
What is the difference between IaaS, PaaS, and SaaS in terms of the control and ownership a consumer has?
Image source: https://www.bmc.com/blogs/saas-vs-paas-vs-iaas-whats-the-difference-and-how-to-choose/
The diagram above exhibits the difference between all the cloud deployment models and the on-premise IT in terms of control and ownership. Following is the list of these deployment models in the descending order of your organization’s control and ability to manage:
Describe the AWS, Azure services for hot storage along with their availability, time to read the first byte, usage charges and best-fit use cases.
In the case of Microsoft Azure, there are few storage services, including Blob storage, file storage, and disk storage. Availability stands at 99.99% while the Time to the first byte is in milliseconds. Most applicable use-cases include where data/file/object storage is required for direct application access. Also, it fits well where the ability to serve the files/static resources to the front-end with performance is of prime importance.
In the case of AWS, the hot storage service is called Amazon S3 Standard. Its availability is 99.99%, and time to the first byte is in milliseconds. Most applicable use-cases include where data/file/object storage is required for direct application access. Also, it fits well where the ability to serve the files/static resources to the front-end with performance is of prime importance.
Describe the AWS, Azure services for cold storage along with their availability, time to read the first byte, usage charges and best-fit use cases.
In the case of Microsoft Azure, there are few cold storage services, such as Blob and general-purpose (GPv2 and GPv1) storage. Availability stands at 99% while the time to the first byte is in milliseconds. It is great for use-cases where there is a lower frequency of data access. Minimum storage duration: 30 days (GPv2). It offers a lower storage cost, but a higher access cost.
In the case of AWS, the S3 Standard-IA is the cold storage service with an availability of 99.9%. Time to the first byte: milliseconds. The most suitable use-cases are where access frequency is low because of its data retrieval fee. Minimum storage duration: 30 days
Explain a container cluster and which cloud services support the deployment of containerized applications on cloud-based clusters?
A container cluster is a dynamic system that manages several containers, groups them together into pods and runs them on nodes. It manages all their interconnections and communication channels. On such a cluster, you can run applications that are modular and can run on multiple containers. A cluster container helps you with a native ability to scale and prevents single points of failure (SPOFs). Examples of container cluster are Docker Swarm and Kubernetes or K8S.
Container clusters are offered by multiple cloud service providers as a service. Azure Container Service (AKS), Elastic Container Service (ECS) by AWS, Google Compute Engine (GKE) by Google Cloud Platform are the most popular container cluster services by the leading cloud service providers.
And that’s a wrap. In addition to the questions, if you wish to do a comprehensive course on the subject that will give you complete knowledge on the subject so that you can answer all the above questions and much more efficiently, visit no other than Manipal Prolearn course on cloud computing.