Cloud Computing (AL-604 (A)) - Unit 5 Short Notes

UNIT 5: Cloud Computing

1. Foundational Concepts & Comparative Analysis

Grid Computing vs. Cloud Computing

Example: SETI@home (Grid) vs. AWS EC2 (Cloud).

Computing on Demand / Utility Computing

• Concept: Provisioning of computing resources (CPU, storage, network) as a metered, on-demand service, analogous to traditional utilities (electricity, water).

• Enabling Characteristics:

  • On-demand self-service: Users provision resources automatically.

  • Elasticity: Resources scale rapidly in/out.

  • Measured service: Resource usage is monitored, controlled, and billed.

• Dynamic Provisioning: Achieved via automation and orchestration (e.g., auto-scaling groups) that respond to predefined metrics (CPU load, network traffic).

Cloud Deployment Models

Selection Framework: Choose based on:

• Cost: Public (OpEx) vs. Private (CapEx).

• Control & Security: Private > Hybrid > Public.

• Compliance: Private/Community for strict regulations.

• Scalability Needs: Public for unpredictable spikes.

Adoption Challenges and Risks

• Business Perspective:

  • Vendor lock-in: Difficulty migrating due to proprietary APIs/services.

  • TCO Uncertainty: Hidden costs (data egress, API calls).

  • Regulatory Compliance: Data sovereignty (GDPR, HIPAA).

• IT Perspective:

  • Integration Complexity: Legacy system integration.

  • Skill Gaps: Need for cloud-native skills (DevOps, serverless).

  • Loss of Direct Control: Over infrastructure and security posture.

2. Cloud Service Models (XaaS)

Control & Responsibility Matrix:

The Cloud Security Alliance (CSA) defines shared responsibility. Provider manages security of the cloud (physical, hypervisor). User manages security in the cloud (OS patching, app config, IAM).

Cloud Stack / Layered Model:

┌─────────────────────────────────────┐

│         Application (SaaS)           │

├─────────────────────────────────────┤

│      Platform (PaaS)                │

├─────────────────────────────────────┤

│    Infrastructure (IaaS)            │

├─────────────────────────────────────┤

│   Virtualization (Hypervisor)       │

├─────────────────────────────────────┤

│        Physical Hardware            │

└─────────────────────────────────────┘

Example: A web app on Heroku (PaaS) uses underlying AWS EC2 (IaaS) and VMware (Virtualization).

3. Virtualization Technologies

Virtualization Fundamentals

• Definition: Abstraction of physical hardware resources (CPU, memory, storage, network) to create multiple isolated virtual machines (VMs) or environments.

• Benefits:

  • Resource Utilization: Higher density (multiple VMs per physical server).

  • Isolation: Fault and security isolation between VMs.

  • Flexibility: Live migration, snapshots, cloning.

Hypervisors / Virtual Machine Monitors (VMMs)

Functions: Resource allocation, VM isolation, hardware emulation, device I/O management.

Server Virtualization & Virtualized Data Centers

• Architecture:

  • Virtualized Servers: Multiple VMs on each physical host.

  • Virtualized Storage: SAN/NAS presented as virtual disks.

  • Virtualized Networking: Virtual switches (vSwitch), VLANs, software-defined networking (SDN).

  • Management Layer: vCenter, OpenStack Nova for orchestration.

DiagramCANVAS: Show physical servers with multiple VMs, connected via virtual switch to virtual storage and network, all managed by a central console.

Storage Virtualization

Logical Partitioning (LPAR)

• Definition: Firmware-level partitioning of a physical server (e.g., IBM POWER, mainframes) into independent logical partitions, each with dedicated CPU, memory, I/O.

• Advantages:

  • Strong isolation (hardware-enforced).

  • Predictable performance (dedicated resources).

  • Security for sensitive workloads.

• Disadvantages:

  • Less flexible than hypervisor-based (static resource allocation).

  • Vendor-specific (IBM, HP).

  • Lower utilization if partitions are under-utilized.

Virtualization Platform Requirements

• Hardware Support: CPU virtualization extensions (Intel VT-x, AMD-V), IOMMU for device assignment.

• Software: 64-bit OS, compatible drivers, management tools (vSphere, libvirt).

4. Cloud Security

Importance & Core Challenges

• Multi-tenancy risks: Data leakage between tenants.

• Loss of control: Physical security managed by provider.

• Shared technology vulnerabilities: Hypervisor attacks, side-channel.

• Compliance: Meeting regulatory standards across jurisdictions.

Security Aspects in Cloud

1. Data Security: Encryption at rest (AES-256), in transit (TLS 1.3), DLP.

2. Network Security: Virtual firewalls (security groups), IDS/IPS, DDoS mitigation (AWS Shield).

3. Identity & Access Management (IAM): Centralized authentication, MFA, RBAC.

4. Compliance & Governance: Audit logs (CloudTrail), standards (ISO 27001, SOC 2).

Secure Execution Environments & Communications

• Encryption: Symmetric (AES) for bulk data, asymmetric (RSA) for key exchange.

• Secure Protocols: TLS/SSL for data in transit, SSH for secure shell.

• Secure Bootstrapping: TPM (Trusted Platform Module) for measured boot, attestation.

Virtual Machine (VM) Security

Recommendations: Use minimal, hardened images; micro-segmentation; continuous monitoring (SIEM); integrity checking (hash verification).

Role-Based Access Control (RBAC)

• Model:

  • Principal: User, group, service.

  • Role: Job function (e.g., Admin, ReadOnly).

  • Permission: Action on resource (e.g., ec2:StartInstances).

• Implementation: AWS IAM Roles, Azure RBAC.

• Principle of Least Privilege: Grant minimum permissions necessary.

Example: A Developer role may have PutObject on S3 but not DeleteBucket.

5. Cloud Architecture & Design Paradigms

Cloud Computing Reference Model

┌─────────────────────────────────────────────┐

│            Front End (Client)               │

│  (Web Browser, Mobile App, Thin Client)    │

├─────────────────────────────────────────────┤

│            Cloud Delivery                   │

│  (SaaS, PaaS, IaaS - Service Models)       │

├─────────────────────────────────────────────┤

│            Back End (Infrastructure)        │

│  (Virtualization, Servers, Storage, Network)│

├─────────────────────────────────────────────┤

│            Management & Security            │

│  (Orchestration, IAM, Monitoring, Billing) │

└─────────────────────────────────────────────┘

Core Components: Front-end (client interface), back-end (cloud resources), cloud delivery (service models), management (provisioning, security).

Service-Oriented Architecture (SOA) in Cloud

• Principles:

  • Loose Coupling: Services independent, communicate via standards.

  • Interoperability: Platform-agnostic (XML, SOAP, REST).

  • Reusability: Services as modular building blocks.

• Facilitating Integration:

  • Web Services/RESTful APIs: Standardized interfaces.

  • Message Queues (e.g., SQS, RabbitMQ): Asynchronous communication.

  • Enterprise Service Bus (ESB): Mediation and routing.

Example: A cloud-based CRM (SaaS) exposes REST API for on-premise ERP integration.

Role of Independent Software Vendors (ISVs)

• Develop applications optimized for cloud platforms (e.g., SAP HANA on AWS).

• Deployment Models:

  • ISV Cloud: ISV operates its own cloud (e.g., Salesforce).

  • Marketplace Distribution: Apps listed on AWS Marketplace, Azure Marketplace.

• Benefits: Faster time-to-market, elastic scaling, managed infrastructure.

Cloud Stack Design

• Mapping: Business requirement → Service model → Technology stack.

• Example (E-Business App):

  • Requirement: Scalable web store with minimal ops.

  • Stack: PaaS (Google App Engine for app code) + IaaS (AWS RDS for database) + SaaS (Stripe for payments).

  • Design: Stateless app tier, managed DB, third-party SaaS for non-core functions.

6. Management, Performance & Tools

Quality of Service (QoS)

• Definition: Measurable service characteristics agreed upon in SLA.

• Key Metrics:

  • Availability: Uptime percentage (e.g., 99.99%).

  • Reliability: Mean Time Between Failures (MTBF).

  • Throughput: Requests per second.

  • Latency: Response time (ms).

• Key Issues:

  • Resource Contention: Noisy neighbor problem in multi-tenant clouds.

  • Performance Variability: Due to shared underlying hardware.

  • SLA Enforcement: Monitoring, penalties, credits.

Cloud Infrastructure Benchmarks

• Performance Metrics: CPU speed (IPS), memory bandwidth, I/O throughput (IOPS), network latency/bandwidth.

• Standard Suites:

  • SPEC Cloud: Comprehensive IaaS benchmarking.

  • YCSB (Yahoo! Cloud Serving Benchmark): NoSQL/DB performance.

• Use: Capacity planning, provider comparison, cost-performance analysis.

Cloud Management & Orchestration Tools

Storage Cloud Considerations

• Object Storage (e.g., S3): HTTP-based, massive scale, eventual consistency, low cost. Use: Static assets, backups.

• Block/File Storage (EBS, Azure Disk): Persistent, POSIX-compliant, high performance. Use: Databases, VM disks.

• Durability vs. Availability: Durability (e.g., 11 9's for S3) ≠ Availability (regional outages).

• Cost Model: Storage cost + API request cost + data transfer cost.

\boxed{\text{Key Exam Focus: Grid vs. Cloud, Deployment Models, Virtualization (Hypervisors, SAN/NAS), Security (VM Risks, RBAC), SOA, QoS, Benchmarks}}