Skip to content

⚡ Cloud & Infrastructure — Points 601–700

Format: Point & Concept → Interview Answer → Example / Tool

Each row gives you a clean definition you can say in an interview, plus a real-world example or tool.


☁️ Cloud Models & Responsibility (601–640)

Point & Concept Interview Answer Example / Tool
601. Cloud Computing Delivering computing services (servers, storage, databases, networks) over the internet "on-demand" with pay-as-you-go pricing. AWS, Azure, Google Cloud (GCP)
602. Shared Responsibility Model A framework dictating that the provider secures the cloud infrastructure, while the customer secures the data and apps in the cloud. AWS secures physical data centres; you secure your S3 bucket
603. IaaS (Infrastructure) The cloud provider gives you the raw virtual hardware (VMs, networking); you manage the OS, runtime, and apps. AWS EC2 or Azure Virtual Machines
604. PaaS (Platform) The provider manages the underlying infrastructure and OS; you just deploy your application code. AWS Elastic Beanstalk or Azure App Service
605. SaaS (Software) The provider manages everything from the infrastructure to the app; you just configure settings and manage users. Microsoft 365, Salesforce, or GitHub
606. Public Cloud Infrastructure shared by multiple customers (tenants) over the public internet. A standard AWS or Azure environment
607. Private Cloud Cloud infrastructure dedicated entirely to a single organisation, hosted on-premise or by a third party. VMware vSphere or an Azure Stack on-premise
608. Hybrid Cloud A mix of on-premise infrastructure and public cloud services, connected by a secure or dedicated link. On-premise Active Directory synced with Entra ID
609. Multi-Cloud Using services from more than one public cloud provider to avoid lock-in or leverage specific features. Using AWS for compute and GCP for BigQuery analytics
610. Cloud Deployment Model The model defining where the infrastructure resides and who has access to it (Public, Private, Hybrid, Community). Moving from a Private Cloud to a Hybrid Deployment
611. Provider Responsibility The physical security of data centres, hardware host patching, and hypervisor isolation. AWS replacing a failed hard drive
612. Customer Responsibility The security of guest OS patching, firewall rules, IAM configurations, and data encryption. Updating Windows on an EC2 instance and configuring Security Groups
613. Cloud Elasticity The ability of a cloud environment to automatically provision and de-provision resources to match fluctuating demand. Auto Scaling Group adding 5 VMs during a Black Friday sale
614. Resource Pooling The provider's ability to serve multiple customers from the same physical hardware using multi-tenancy. Ten different customer VMs running securely on one physical AWS host
615. Broad Network Access Cloud services are accessible from anywhere via standard internet protocols and devices. Logging into the Azure Portal from a laptop over HTTPS
616. Serverless Computing A cloud execution model where the provider dynamically manages the allocation and provisioning of servers; the customer only pays for the execution time of their code. AWS Lambda or Azure Functions executing a piece of code upon an API request
617. Containers Lightweight, standalone, executable packages of software that include everything needed to run an application. A Docker container running an Nginx web server
618. Kubernetes (K8s) An open-source system for automating the deployment, scaling, and management of containerised applications. A K8s cluster managing the lifecycle of 50 Docker containers
619. Container Registry A repository for storing and distributing container images securely. AWS ECR, Docker Hub, or Azure Container Registry
620. Ephemeral Compute Virtual machines or containers that are spun up dynamically to perform a task and destroyed when the task completes. Spot instances processing a batch job and shutting down
621. Cloud Native Building applications specifically designed to leverage cloud computing architectures like microservices and serverless. A web app built with Lambda functions and DynamoDB
622. Microservices An architectural style that structures an application as a collection of loosely coupled, independently deployable services. A shopping app where the cart, payment, and inventory are separate services
623. Infrastructure as Code (IaC) Managing and provisioning data centres through machine-readable definition files, rather than physical hardware configuration. Terraform creating 10 VMs and a load balancer with a single script
624. Immutable Infrastructure An approach to managing services where components
are replaced rather than changed once they have been deployed. Triggering a Terraform script to deploy a new server and destroying the old server instead of updating the existing one
625. Cloud Lock-in The difficulty of moving an application or data from one cloud provider to another due to dependence on proprietary services. Being unable to easily migrate off AWS Lambda due to heavy use of AWS Step Functions
626. High Availability (HA) in Cloud Designing infrastructure across multiple physically separate locations to ensure continuous operation. Deploying web servers across three AWS Availability Zones (AZs)
627. Availability Zone (AZ) One or more discrete data centres with redundant power, networking, and connectivity within a cloud region. AWS us-east-1a
628. Cloud Region A geographic area containing two or more Availability Zones. Azure UK South (London)
629. Edge Computing Processing data near the edge of your network, where the data is being generated, instead of in an artificial intelligence cloud. An IoT device doing local video analysis before sending summaries to AWS
630. CDN (Content Delivery Network) A geographically distributed network of proxy servers and their data centres. Cloudflare caching images in Sydney to serve Australian users faster
631. Cloud Firewalls Virtualised network security devices that control traffic to and from cloud resources. AWS Security Groups acting as stateful firewalls per EC2 instance
632. CASB (Cloud Access Security Broker) Software that sits between users and cloud applications to monitor activity, enforce policy, and prevent data leakage. Microsoft Defender for Cloud Apps blocking uploads of PII to Box
633. CSPM (Cloud Security Posture Management) Tools that continuously monitor cloud configurations to identify misconfigurations, compliance violations, and risks. Prisma Cloud detecting an S3 bucket that was accidentally made public
634. CWPP (Cloud Workload Protection Platform) Security tools focused on protecting the compute workloads (VMs, containers, serverless) from compromise. SentinelOne deployed on an EC2 instance to stop malware execution
635. CNAPP (Cloud Native Application Protection Platform) An architectural approach that combines CSPM, CWPP, and CIEM into a single platform for holistic cloud security. Wiz or Orca Security analyzing both configuration and workload vulnerabilities
636. CIEM (Cloud Infrastructure Entitlement Management) Tools that manage and monitor identities and privileges across multi-cloud environments to enforce least privilege. Finding inactive IAM roles in AWS and removing their permissions
637. Cloud IAM Identity and Access Management specifically controlling access to cloud provider APIs and resources. AWS IAM assigning a role to an EC2 instance to let it read from S3
638. Cloud API Security Protecting the application programming interfaces that developers and services use to interact with cloud infrastructure. Requiring TLS and API keys for all requests to AWS API Gateway
639. VPC (Virtual Private Cloud) A logically isolated section of the public cloud where you can launch resources in a virtual network you define. Creating a production VPC in Azure with its own subnets and routing
640. Cloud Compliance Frameworks Pre-defined sets of rules and controls (like SOC 2, HIPAA, FedRAMP) applied to cloud environments to meet regulatory standards. Using AWS Artifact to download SOC 2 compliance reports

☁️ Advanced Cloud Security & DevSecOps (641–700)

Point & Concept Interview Answer Example / Tool
641. Instance Role / Managed Identity A way to grant a cloud virtual machine permissions to access other cloud resources securely, without hardcoding credentials. An EC2 instance using an IAM Role to securely write logs to CloudWatch
642. Secrets Management The secure storage, rotation, and distribution of sensitive credentials (API keys, passwords, certificates) avoiding hardcoding them in code. AWS Secrets Manager or HashiCorp Vault injecting a DB password at runtime
643. Key Management Service (KMS) A managed service that allows you to create and control the cryptographic keys securely used to protect your data. Azure Key Vault holding the master key used to encrypt a database
644. Cloud Trail / Audit Logging A service that records API activity and user actions within a cloud environment to enable governance and compliance. AWS CloudTrail logging which user deleted a specific DynamoDB table
645. S3 Bucket Takeover / Dangling DNS An attack where a domain points to a cloud resource (like an S3 bucket) that was deleted; an attacker recreates the bucket to hijack the domain. A company deletes images.mycompany.com S3 bucket but forgets to delete the DNS record
646. Control Plane vs Data Plane The Control Plane manages the architecture (e.g. creating VMs); the Data Plane is the actual user data flowing through the architecture. Control Plane: aws ec2 run-instances. Data Plane: Users visiting the resulting web server.
647. Metadata Service (IMDS) An API endpoint available from within a cloud VM used to retrieve instance metadata, including temporary IAM credentials. Querying 169.254.169.254 from an EC2 instance to get its temporary access token
648. SSRF (Server-Side Request Forgery) in Cloud A vulnerability where an attacker forces a server to make a request to the cloud metadata service to steal credentials. Capital One breach (2019) where an SSRF vulnerability was used to access the AWS metadata service
649. S3 Public Exposure One of the most common cloud breaches where a storage bucket's permissions permit unauthorized Internet access to files. Leaving an S3 bucket containing customer backups open to the world
650. IAM Over-Privilege The risky practice of granting more permissions than necessary, often using wildcard * permissions like AdministratorAccess. A developer granting s3:* to a Lambda function instead of s3:GetObject on a specific bucket
651. Cross-Account Access The ability to securely grant users in one cloud account access to resources in another without sharing credentials. Using AWS AssumeRole to let the central Security Account audit a Production Account
652. Security Group A fundamental cloud firewall operating at the instance/VM level, blocking traffic by default and allowing specific inbound/outbound ports. Allowing TCP port 22 (SSH) only from the corporate office IP
653. Network ACL (NACL) A stateless firewall operating at the subnet level in a VPC, providing an additional layer of defence. Denying all traffic from a known malicious IP range entering the subnet
654. Cloud Load Balancer Logging Enabling logs on the load balancer to capture the source IP, path, and user agent of every single client request before it hits internal servers. ALB Access Logs sent to an S3 bucket for incident response analysis
655. DevSecOps The integration of security practices and automated testing within the Dev/Ops software development lifecycle (SDLC). A CI/CD pipeline that runs a vulnerability scan before deploying code
656. SAST (Static Application Security Testing) Analysing application source code (without executing it) to find security vulnerabilities early in development. Using Checkmarx or SonarQube to find SQL injection flaws in a Java repository
657. DAST (Dynamic Application Security Testing) Analysing a running web application from the outside by sending simulated attacks to find vulnerabilities. Running OWASP ZAP or Burp Enterprise against a staging environment
658. SCA (Software Composition Analysis) Identifying third-party open-source libraries in your codebase and checking them against databases of known vulnerabilities. Using Snyk or Dependabot to find out you're using a vulnerable version of Log4j
659. Vulnerability Scanning (Containers) Inspecting container images for known CVEs before they are pushed to production or run in a cluster. Trivy scanning a Docker image in a GitHub Action
660. Secrets Scanning Automatically reviewing code commits to detect hardcoded passwords, API keys, or cloud access tokens before they are merged. GitGuardian blocking a commit that contains an AWS API secret
661. Image Signing Cryptographically signing container images so the orchestration platform (like K8s) can verify they haven't been tampered with. Docker Content Trust ensuring that the cluster only runs approved, signed images
662. Immutable Infrastructure Deploying servers or containers that are never modified after deployment. If an update is needed, a completely new instance is built and deployed. Stopping SSH access entirely; replacing EC2 instances via Terraform instead of updating them
663. Golden Image An OS image pre-configured with the company's security baseline, agents, and patches from which all new servers must be built. Creating an AMI (Amazon Machine Image) that has CrowdStrike and Splunk pre-installed
664. Drift Detection Continuously monitoring infrastructure to ensure its actual configuration matches the expected Infrastructure-as-Code state. AWS CloudFormation drift detection showing someone manually opened port 22
665. Shift Left The philosophy of moving security testing and controls as early as possible in the software development lifecycle to catch issues sooner and cheaper. Implementing SAST in the developer's IDE instead of waiting for a pentest in staging
666. Policy as Code Writing security and compliance rules in a programming language to automatically evaluate and block non-compliant infrastructure changes. Using Open Policy Agent (OPA) to reject Terraform plans that create public S3 buckets
667. CI/CD Pipeline Security Defending the automated build and deployment systems against compromise, as gaining access to the pipeline means gaining access to production. Using short-lived OpenID Connect (OIDC) tokens instead of long-lived API keys in GitHub Actions
668. Cloud Storage Encryption Configuring object and block storage services to encrypt data at rest, often managed by the cloud provider's Key Management Service. Enabling AWS S3 Default Encryption using SSE-KMS
669. Service Mesh A dedicated infrastructure layer for managing service-to-service communication within a microservices architecture, often enforcing mutual TLS. Istio managing internal communication rules between K8s pods
670. API Gateway security Centralising security for APIs by enforcing OAuth validation, rate limiting, and input filtering before traffic reaches backend systems. AWS API Gateway blocking a brute-force attack with a strict request quota
671. Cloud Data Loss Prevention (DLP) Tools that scan data moving into or already living in cloud storage to identify and protect sensitive information such as PII. Amazon Macie discovering unencrypted credit card numbers in an S3 bucket
672. Bastion Host A highly secured, heavily monitored server placed in a public subnet that administrators use to connect to instances in private subnets. An EC2 instance used exclusively as a jump box for SSH access
673. Route Tables Rules within a VPC that determined where network traffic from your subnet or gateway is directed. A route table directing 0.0.0.0/0 (internet bound traffic) to an Internet Gateway
674. Egress Filtering Restricting outward-bound network traffic from cloud servers to prevent malware from downloading extra payloads or exfiltrating data. Creating a Security Group rule that only allows outbound traffic on port 443
675. Zero Trust Architecture The principle of "never trust, always verify"; requiring continuous authentication and authorisation for all users and devices regardless of location. BeyondCorp replacing enterprise VPNs with verified context-based access
676. Serverless Security The challenge of securing functions-as-a-service where you don't control the OS. Security focuses heavily on IAM permissions and application code. Writing strict IAM roles for AWS Lambda
677. Cloud Auditing Periodically verifying cloud configurations, access logs, and security controls against industry standards like CIS. Using AWS Security Hub to run automated CIS benchmark checks
678. Ephemeral Credentials Temporary, short-lived security tokens used to access cloud resources instead of long-term static passwords or keys. AWS STS generating access tokens that expire automatically after one hour
679. Incident Response in Cloud Customising traditional IR processes to account for cloud environments, relying heavily on automation, snapshots, and API logs. Using AWS Step Functions to automatically isolate a compromised EC2 instance
680. Cloud Vendor Lock-In Risk The operational risk of relying too heavily on provider-specific proprietary services, making migration technically or financially prohibitive. Architecting with generic Kubernetes instead of provider-specific PaaS
681. Bring Your Own Key (BYOK) A model where a customer generates their own encryption key on-premise and imports it to the cloud provider's KMS. Importing an RSA key to Azure Key Vault to encrypt a database
682. Customer Managed Key (CMK) Encryption keys within the cloud provider's KMS where the customer retains full control over the lifecycle, rotation, and usage policies. Creating an AWS KMS CMK with a strict key policy limiting usage to a single IAM role
683. WAF (Web Application Firewall) A cloud-hosted appliance that inspects incoming HTTP/HTTPS traffic to block common attacks like SQL injection and Cross-Site Scripting. AWS WAF deployed on an Application Load Balancer dropping malicious requests
684. Cloud Threat Intelligence Feeds of known malicious IP addresses, domains, and files integrated into cloud security tools to automatically block attacks. Amazon GuardDuty flagging traffic communicating with a known crypto-mining pool
685. Data Sovereignty Legal guidelines stipulating that data must reside physically within the borders of the country where it was generated. Configuring Azure to only store European customer data within the Frankfurt region
686. RTO (Recovery Time Objective) The maximum acceptable amount of time that an application can be offline during an outage. A disaster recovery plan designed to restore services within 4 hours
687. RPO (Recovery Point Objective) The maximum acceptable amount of data loss measured in time before a failure occurs. 15-minute database backups ensuring the RPO is less than 15 minutes
688. Snapshot A point-in-time copy of a cloud storage volume that can be used to quickly restore data or create a duplicate environment. Taking an EBS snapshot before applying a major server patch
689. AWS GuardDuty An AWS threat detection service that continuously monitors CloudTrail logs, VPC Flow Logs, and DNS logs for malicious activity. GuardDuty alerting you that an EC2 instance is port scanning internal networks
690. Azure Security Center Microsoft's unified infrastructure security management system that strengthens the security posture of data centres. Identifying VMs missing endpoint protection agents
691. GCP Security Command Center Google Cloud's centralised vulnerability and threat reporting service. Identifying misconfigured Google Cloud Storage buckets
692. Least Privilege in Cloud Ensuring that users, groups, and automated services have precisely the permissions needed to perform their tasks, and nothing more. Giving a user dynamodb:query access to a single table, not the whole database
693. Infrastructure as Code Scanning Automatically reviewing Terraform, CloudFormation, or Ansible code before deployment to catch misconfigurations. Checkov blocking a Terraform plan because it creates a database without encryption
694. Container Escape An attack where an adversary uses an exploit within a container to gain access to the underlying host operating system. Exploiting a misconfigured privilege container to access the host's file system
695. RBAC Role-Based Access Control, mapping permissions to roles (like 'Developer' or 'Admin') rather than individuals. Assigning a new employee to the 'Data Analyst' role in Azure Active Directory
696. Multi-tenant Architecture A single instance of software running on a server that serves multiple tenants (customers), reliant on strict logical separation. Salesforce hosting thousands of clients securely in the same application stack
697. Egress Data Charges The cost cloud providers charge you for data leaving their network, often a massive and unexpected expense during large-scale exfiltration or DDoSes. An attacker exfiltrating 5TB over a weekend causing a massive cloud bill shock
698. Shadow IT in Cloud When employees provision SaaS accounts or unmanaged cloud servers without the IT department's knowledge or approval. A marketing team buying a standalone Google Workspace account with a corporate credit card
699. Cloud Cost Optimisation The ongoing process of ensuring you only pay for the cloud resources you actually need, often driven by tagging and automation. Ensuring non-production development environments automatically shut down over the weekend
700. Cloud Resource Tagging Applying metadata (labels) to cloud resources to help identify their owner, environment, or purpose, aiding both billing and security incident response. Adding the tag Environment=Production and Owner=SecurityTeam to a firewall