| Model | Definition | You Manage | Provider Manages | Examples | Short Interview Answer |
|---|---|---|---|---|---|
| IaaS (Infrastructure as a Service) |
Provides virtualized infrastructure resources. | OS, Runtime, Application, Data | Servers, Storage, Network, Virtualization |
AWS EC2 Azure VM Google Compute Engine |
"IaaS provides virtual machines and infrastructure where users manage the OS and applications." |
| PaaS (Platform as a Service) |
Provides platform for developing and deploying applications. | Application, Data | Infrastructure, OS, Runtime, Scaling |
Azure App Service AWS Elastic Beanstalk OpenShift |
"PaaS allows developers to deploy applications without managing servers or operating systems." |
| SaaS (Software as a Service) |
Ready-to-use software delivered over the internet. | Only Usage | Everything |
Gmail Salesforce Microsoft 365 |
"SaaS provides fully managed software that users access through a browser." |
| Model | Definition | Characteristics | Examples | Short Interview Answer |
|---|---|---|---|---|
| Public Cloud | Cloud infrastructure shared among multiple organizations. |
Pay-as-you-go Highly scalable Internet accessible |
AWS Azure GCP |
"Public cloud provides shared infrastructure managed by a cloud provider and is ideal for scalable applications." |
| Private Cloud | Cloud environment dedicated to a single organization. |
High security Full control Higher cost |
On-Prem Data Center Internal OpenShift Cluster |
"Private cloud offers dedicated infrastructure with greater security and compliance control." |
| Hybrid Cloud | Combination of Public and Private Cloud. |
Flexibility Cost optimization Better disaster recovery |
On-Prem Database + Azure AKS Private DC + AWS Services |
"Hybrid cloud combines private and public cloud environments to leverage the benefits of both." |
You Manage:
Cloud Provider Manages:
java -jar app.jar
You manually install Java, configure server, deploy application and maintain OS.
You Manage:
Provider Manages:
Deploy Spring Boot application and platform handles:
You simply consume software.
| Feature | Public Cloud | Private Cloud | Hybrid Cloud |
|---|---|---|---|
| Ownership | Cloud Provider | Single Organization | Both |
| Cost | Lower | Higher | Medium |
| Scalability | High | Limited | High |
| Security | Shared Responsibility | Full Control | Balanced |
A geographical area containing multiple data centers.
Examples:Independent data centers within a region.
Example:Region: AWS Mumbai AZs: ap-south-1a ap-south-1b ap-south-1c
4 GB RAM VM ? 16 GB RAM VM
Limitation: Hardware limits.
1 Pod ? 10 Pods
Automatically scale resources based on demand.
Example:Normal Traffic = 2 Pods Peak Traffic = 10 Pods Night Time = 2 Pods
System remains operational even if some components fail.
User Service Replica-1 Replica-2 Replica-3 If Replica-1 crashes, Replica-2 and Replica-3 continue serving requests.
Ability to continue functioning despite failures.
Payment Service Down Retry 3 Times ? Fallback Response ? Log Failure ? Alert Support Team
One application serving multiple customers (tenants).
| Approach | Isolation | Cost |
|---|---|---|
| Shared DB, Shared Schema | Low | Low |
| Shared DB, Separate Schema | Medium | Medium |
| Separate DB per Tenant | High | High |
| Term | Definition |
|---|---|
| Elasticity | Automatic scaling based on demand. |
| Scalability | Ability to increase capacity. |
| High Availability | Maximum uptime with minimal downtime. |
| Fault Tolerance | Ability to survive failures. |
| Region | Geographical cloud location. |
| AZ | Independent data center inside a region. |
| Multi-Tenancy | Single application serving multiple customers. |
Answer:
Horizontal scaling provides better fault tolerance, load distribution, high availability, and supports cloud-native architectures. Vertical scaling is limited by hardware capacity.
Answer:
Answer:
Applications deployed across multiple AZs remain available even if one data center fails. Traffic automatically shifts to healthy AZs.
Answer:
Answer:
A software architecture where a single application instance serves multiple customers while keeping tenant data isolated.
Answer:
Cloud provider secures infrastructure, while customers secure applications, data, access controls, and configurations.
Answer:
| High Availability | Fault Tolerance |
|---|---|
| Minimizes downtime | Continues operating despite failure |
| Recovery after failure | No interruption during failure |
Answer:
Auto Scaling automatically adds or removes resources based on CPU, memory, request count, or custom metrics.
Answer:
Sample Answer:
We deployed Spring Boot microservices on AKS/OpenShift, used Azure Load Balancer, Azure SQL/PostgreSQL, Redis caching, Kafka messaging, Prometheus/Grafana monitoring, and configured HPA for automatic scaling. Applications were deployed across multiple availability zones to achieve high availability and fault tolerance.