AWS and Cloud Computing Interview Questions India 2026: EC2, S3, Lambda, and Cloud Architecture

AWS core service interview questions:

EC2 (Elastic Compute Cloud): 1. 'What is the difference between an EC2 instance store and an EBS volume?' Instance store: physically attached to the host machine; faster I/O; data is lost when the instance stops or terminates (ephemeral). EBS (Elastic Block Store): network-attached block storage; persists independently of the instance lifecycle; can be detached and reattached; use for operating system volumes and databases.

2. 'What are the EC2 pricing models?' On-Demand: pay per second or per hour; no commitment; highest per-unit cost. Reserved Instances: 1-year or 3-year commitment; 40-60% discount versus On-Demand. Spot Instances: unused AWS capacity at up to 90% discount; can be interrupted with 2-minute notice; use for stateless batch processing and fault-tolerant workloads. Savings Plans: flexible discount in exchange for committing to a consistent amount of usage (measured in $/hour); covers any instance type or region.

S3 (Simple Storage Service): 3. 'What is an S3 bucket lifecycle policy?' Automates moving objects between storage classes or deleting them after a defined period. Example: move objects to S3 Standard-IA after 30 days; move to S3 Glacier after 90 days; delete after 365 days. Reduces storage cost without manual intervention.

4. 'What is S3 versioning and why would you enable it?' Keeps multiple versions of each object. Protects against accidental deletion (deletion adds a delete marker; the previous version remains). Use with S3 Object Lock for compliance (immutable copies for financial records, legal documents).

Lambda (Serverless): 5. 'What is a Lambda cold start and how do you mitigate it?' Cold start: the first invocation after a period of inactivity incurs additional latency (100ms-2 seconds) because AWS must initialise a new execution environment. Mitigation: provisioned concurrency (Lambda keeps N instances warm; eliminates cold starts at extra cost); reduce deployment package size and initialisation code; use Lambda SnapStart for Java (pre-initialises the JVM).

AWS networking and security interview questions:

VPC (Virtual Private Cloud): 1. 'What is the difference between a Security Group and a Network ACL?' Security Group: stateful (if you allow inbound traffic on port 80, the return traffic is automatically allowed); operates at the instance level; allow rules only (no explicit deny). Network ACL: stateless (you must allow both inbound and outbound explicitly); operates at the subnet level; both allow and deny rules; evaluated in number order (lower number = higher priority).

2. 'What is a NAT Gateway?' Allows instances in a private subnet to initiate outbound internet connections (software updates, API calls) without being directly reachable from the internet. The NAT Gateway lives in a public subnet; private subnet has a route to the NAT Gateway for 0.0.0.0/0 traffic.

IAM (Identity and Access Management): 3. 'What is the difference between an IAM User and an IAM Role?' IAM User: a person or application with permanent long-term credentials (access key + secret key). IAM Role: an identity with temporary credentials assumed by services, EC2 instances, Lambda functions, or even IAM Users from other accounts. Always use IAM Roles for EC2 instances and Lambda functions instead of embedding access keys in the code or environment variables.

4. 'What is IAM least privilege?' Grant only the permissions needed to perform the task. Do not use the root account for day-to-day operations. Do not attach AdministratorAccess to application roles. Use resource-level restrictions in policies (restrict S3 access to the specific bucket, not 'arn:aws:s3:::*').

RDS and DynamoDB: 5. 'When would you choose DynamoDB over RDS?' DynamoDB: NoSQL, key-value/document model; single-digit millisecond latency at any scale; serverless (no instance management); ideal for high-throughput, predictable-access-pattern workloads (session stores, shopping carts, leaderboards). RDS: relational model (MySQL, PostgreSQL, Aurora); complex joins and transactions; ACID compliance; ideal for structured data with complex relationships (order management, financial ledgers).

Cloud architecture interview questions:

1. 'How would you design a highly available web application on AWS?' Multi-AZ architecture: deploy EC2 instances (or containers on ECS/EKS) across at least 2 Availability Zones. Application Load Balancer (ALB): routes traffic to healthy instances; performs health checks; removes unhealthy instances from rotation. Auto Scaling Group (ASG): automatically adds or removes EC2 instances based on CPU/request count; minimum 2 instances across 2 AZs for HA. RDS Multi-AZ: synchronous standby replica in a second AZ; automatic failover in case of primary failure (< 2 minutes). CloudFront CDN: cache static assets at edge locations globally; reduces latency for Indian users accessing a globally deployed app. Route 53: DNS with health checks; automatic failover to a secondary region if the primary region becomes unavailable.

2. 'What is the difference between horizontal scaling and vertical scaling? When would you use each?' Horizontal scaling (scale out): add more instances. Used for stateless services (web servers, API servers) where load can be distributed. Easier to achieve high availability. Vertical scaling (scale up): increase the instance size (more CPU, more RAM). Used for stateful workloads (databases) where you cannot easily add more instances. Has a ceiling (the largest available instance size). Preferred approach in cloud: design stateless applications so you can scale horizontally; scale databases vertically until you need to shard.

3. 'What is a CDN and how does it improve performance for Indian users?' CDN (Content Delivery Network): a distributed network of edge servers that cache content close to the user. AWS CloudFront has edge locations in Mumbai, Chennai, Bangalore, Hyderabad, Delhi, and Kolkata. Benefits for Indian users: static assets (JS, CSS, images) load from a Mumbai edge server (10-20ms) instead of a US origin server (200-300ms). Reduces origin server load: 70-90% of static asset traffic can be served from the CDN cache.