💻Parallel and Distributed Computing Unit 14 – Cloud Computing & Virtualization

Key Concepts and Definitions

• Cloud computing delivers computing services over the internet (the cloud) including servers, storage, databases, networking, software, analytics, and intelligence
• Virtualization abstracts physical hardware, creating a virtual computing environment where resources can be efficiently utilized and dynamically allocated
• Scalability enables a system to handle increased loads by dynamically adding more resources (horizontal scaling) or increasing the capacity of existing resources (vertical scaling)
• Elasticity allows resources to be rapidly provisioned and released based on demand, ensuring optimal resource utilization and cost-efficiency
• Service Level Agreement (SLA) defines the level of service expected from a cloud provider, including uptime, performance, and security commitments
• Multitenancy enables multiple customers (tenants) to share the same computing resources while maintaining data isolation and security
• Workload refers to the amount of processing a computer system performs, which can be distributed across multiple cloud resources for efficient execution

Evolution of Cloud Computing

• Mainframe computing (1950s-1970s) centralized computing resources, accessed through terminals, laying the foundation for modern cloud computing concepts
• Client-server model (1980s-1990s) distributed computing tasks between clients and servers, enabling resource sharing and remote access
• Grid computing (1990s-2000s) connected geographically distributed computers to solve complex problems, introducing the concept of resource pooling
• Utility computing (early 2000s) provided computing resources as a metered service, similar to electricity or water, paving the way for pay-as-you-go cloud services
• Cloud computing (2006-present) emerged with Amazon Web Services (AWS), followed by other providers like Google Cloud Platform and Microsoft Azure
  • Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS) models were introduced, offering different levels of abstraction and control
• Serverless computing (2014-present) enables developers to focus on writing code without managing underlying infrastructure, with providers dynamically allocating resources as needed

Cloud Service Models

• Infrastructure as a Service (IaaS) provides virtualized computing resources over the internet, including servers, storage, and networking
  • Customers have control over operating systems, storage, and deployed applications, while the cloud provider manages the underlying infrastructure
  • Examples include Amazon EC2, Google Compute Engine, and Microsoft Azure Virtual Machines
• Platform as a Service (PaaS) offers a complete development and deployment environment in the cloud, including tools for building, testing, and deploying applications
  • Developers can focus on writing code without worrying about infrastructure management or software updates
  • Examples include Google App Engine, Microsoft Azure App Service, and Heroku
• Software as a Service (SaaS) delivers software applications over the internet, accessible through a web browser or API
  • Users can access the software from any device with an internet connection, without the need for installation or maintenance
  • Examples include Salesforce, Google Workspace (formerly G Suite), and Microsoft Office 365
• Function as a Service (FaaS) allows developers to execute individual functions in response to events or requests, without managing servers or infrastructure
  • Cloud providers dynamically allocate resources and charge based on the number of function executions and their duration
  • Examples include AWS Lambda, Google Cloud Functions, and Microsoft Azure Functions

Virtualization Technologies

• Hypervisor (virtual machine monitor) creates and manages virtual machines (VMs), allowing multiple operating systems to run concurrently on a single physical machine
  • Type 1 hypervisors (bare-metal) run directly on the host's hardware, providing better performance and security (Xen, Microsoft Hyper-V, VMware ESXi)
  • Type 2 hypervisors (hosted) run as a software layer on top of an existing operating system, offering flexibility and ease of use (Oracle VirtualBox, VMware Workstation)
• Virtual machines (VMs) emulate physical computers, with their own virtual hardware (CPU, memory, storage) and operating system, isolated from other VMs and the host
• Containers provide a lightweight alternative to VMs, packaging an application and its dependencies into a single unit that can run consistently across different environments
  • Containers share the host operating system kernel, resulting in faster startup times and lower overhead compared to VMs
  • Examples include Docker, Kubernetes, and Linux Containers (LXC)
• Software-defined networking (SDN) decouples network control from the underlying hardware, enabling programmatic management and dynamic configuration of network resources
• Storage virtualization abstracts physical storage devices, presenting them as a single logical unit, facilitating data management, backup, and disaster recovery

Cloud Infrastructure and Architecture

• Data centers house the physical infrastructure (servers, storage, networking) that powers cloud computing services, designed for reliability, security, and efficiency
• Compute resources include virtual machines (VMs), containers, and serverless computing options, provisioned on-demand to meet application requirements
• Storage services offer various options for storing and accessing data in the cloud, such as object storage (Amazon S3), block storage (Amazon EBS), and file storage (Google Cloud Filestore)
  • Storage tiers (hot, warm, cold) cater to different access frequency and latency requirements, optimizing costs based on data usage patterns
• Networking infrastructure connects cloud resources and enables communication between services, typically using software-defined networking (SDN) for flexibility and scalability
  • Virtual private cloud (VPC) isolates cloud resources within a logically separated network, ensuring security and control over network configuration
• Load balancing distributes incoming traffic across multiple resources (servers, VMs, containers) to optimize performance, ensure high availability, and handle failover scenarios
• Content delivery networks (CDNs) cache static content in geographically distributed edge locations, reducing latency and improving user experience for globally accessed applications
• API gateways act as a single entry point for managing, securing, and monitoring API requests, facilitating communication between services and enabling serverless architectures

Deployment Models and Strategies

• Public cloud offers computing resources and services over the public internet, owned and operated by a third-party provider (AWS, Google Cloud, Microsoft Azure)
  • Resources are shared among multiple tenants, with the provider responsible for infrastructure management, security, and maintenance
  • Offers high scalability, flexibility, and cost-efficiency, with a pay-as-you-go pricing model
• Private cloud delivers cloud computing services exclusively for a single organization, either on-premises or hosted by a third-party provider
  • Provides greater control, security, and customization compared to public cloud, but requires significant upfront investment and ongoing maintenance
  • Suitable for organizations with strict regulatory compliance or data sovereignty requirements
• Hybrid cloud combines public and private cloud environments, allowing workloads to move between them based on specific requirements (cost, performance, security)
  • Enables organizations to leverage the scalability of public cloud while keeping sensitive data or critical workloads in a private cloud environment
  • Requires careful planning and management to ensure seamless integration and data synchronization between the two environments
• Multi-cloud strategy involves using services from multiple cloud providers to avoid vendor lock-in, optimize costs, and leverage best-of-breed services
  • Requires expertise in managing and integrating different cloud platforms, as well as ensuring data portability and interoperability
• Lift and shift (rehosting) migration strategy involves moving existing applications to the cloud without significant modifications, minimizing initial effort but potentially limiting cloud benefits
• Refactoring (rearchitecting) migration strategy involves redesigning applications to take full advantage of cloud-native features and services, maximizing benefits but requiring more effort and resources

Security and Privacy Considerations

• Data encryption protects sensitive information both at rest (stored) and in transit (during transmission) using encryption algorithms and key management systems
  • Client-side encryption ensures data is encrypted before being sent to the cloud, with the keys held by the customer for added security
  • Server-side encryption is performed by the cloud provider, offering convenience but requiring trust in the provider's security practices
• Access control mechanisms (IAM) regulate who can access cloud resources and what actions they can perform, based on the principle of least privilege
  • Multi-factor authentication (MFA) adds an extra layer of security by requiring users to provide additional verification (e.g., a code from a mobile app) during login
  • Role-based access control (RBAC) assigns permissions to users based on their roles and responsibilities within the organization
• Network security measures protect cloud resources from unauthorized access and potential threats, using tools like firewalls, intrusion detection/prevention systems (IDS/IPS), and virtual private networks (VPNs)
• Compliance with industry standards and regulations (HIPAA, GDPR, PCI-DSS) ensures that cloud services meet specific security and privacy requirements for sensitive data and applications
• Shared responsibility model defines the division of security responsibilities between the cloud provider and the customer, depending on the service model (IaaS, PaaS, SaaS)
  • The provider is responsible for securing the underlying infrastructure, while the customer is responsible for securing their applications, data, and user access
• Security monitoring and incident response processes help detect, investigate, and mitigate security breaches or anomalies in the cloud environment
  • Cloud providers offer native security tools and services (AWS GuardDuty, Google Cloud Security Command Center) for continuous monitoring and threat detection
  • Third-party security solutions can be integrated to enhance visibility and protection across multiple cloud platforms and on-premises systems

Performance and Scalability

• Auto-scaling dynamically adjusts the number of resources (VMs, containers) based on predefined rules or metrics (CPU utilization, request rate) to handle fluctuating workloads
  • Horizontal scaling (scaling out) adds more instances to distribute the load, suitable for stateless applications or those that can be easily parallelized
  • Vertical scaling (scaling up) increases the capacity of existing instances (CPU, memory), suitable for stateful applications or databases that require more resources
• Load balancing distributes incoming traffic across multiple resources to optimize performance, ensure high availability, and handle failover scenarios
  • Application load balancers (ALBs) operate at the application layer (HTTP/HTTPS), routing traffic based on content and supporting advanced features like sticky sessions and SSL termination
  • Network load balancers (NLBs) operate at the transport layer (TCP/UDP), providing high-speed, low-latency traffic distribution for non-HTTP applications or those requiring high throughput
• Caching stores frequently accessed data or computations in fast, temporary storage (memory, SSDs) to reduce latency and improve application performance
  • In-memory caches (Redis, Memcached) store data in RAM for extremely fast access, ideal for session management, real-time analytics, or content delivery
  • Content delivery networks (CDNs) cache static content in geographically distributed edge locations, reducing latency for users accessing the application from different regions
• Serverless computing (FaaS) automatically scales resources based on the number of incoming requests or events, allowing developers to focus on writing code without managing infrastructure
  • Serverless platforms (AWS Lambda, Google Cloud Functions) handle the provisioning, scaling, and management of the underlying resources, charging only for the actual execution time and resources consumed
• Performance monitoring and optimization tools help identify bottlenecks, analyze resource utilization, and fine-tune application performance in the cloud
  • Cloud providers offer native monitoring solutions (Amazon CloudWatch, Google Cloud Monitoring) for collecting and visualizing metrics, logs, and events
  • Third-party application performance management (APM) tools (New Relic, Datadog) provide deeper insights into application behavior, user experience, and performance anomalies

Real-World Applications and Case Studies

• Netflix leverages AWS for its global streaming service, using a microservices architecture, auto-scaling, and content delivery networks (CDNs) to deliver high-quality video to millions of users
  • Netflix uses AWS Lambda for serverless computing, Amazon S3 for storage, and Amazon DynamoDB for its distributed database needs
  • The company's cloud-native approach enables rapid innovation, scalability, and resilience, ensuring a seamless user experience across different devices and regions
• Airbnb migrated its infrastructure to AWS to handle its rapid growth and global expansion, using a mix of IaaS, PaaS, and SaaS services
  • Airbnb uses Amazon EC2 for compute, Amazon RDS for managed databases, and Amazon ECS for container orchestration
  • The company also leverages machine learning services (Amazon SageMaker) for personalized recommendations and fraud detection, improving user experience and trust
• Spotify transitioned from on-premises data centers to Google Cloud Platform (GCP) to scale its music streaming service and leverage advanced data analytics capabilities
  • Spotify uses Google Compute Engine for compute, Google Cloud Storage for storing audio files, and Google Cloud Dataproc for big data processing (Hadoop, Spark)
  • The company also employs machine learning (TensorFlow) and data analytics (BigQuery) to personalize playlists, recommend songs, and gain insights into user behavior
• The New York Times used Google Cloud to digitize its entire archive of over 5 million historical articles, making them searchable and accessible to readers worldwide
  • The company used Google Cloud Storage for storing the scanned images and OCR data, and Google Cloud Dataflow for parallel processing and indexing of the articles
  • The project showcased the power of cloud computing in preserving and democratizing access to historical information, enabling new research and discovery opportunities
• Coca-Cola Enterprises migrated its global IT infrastructure to Microsoft Azure, leveraging a hybrid cloud approach for its mission-critical applications and data
  • The company uses Azure Virtual Machines for compute, Azure SQL Database for managed databases, and Azure Active Directory for identity and access management
  • The migration helped Coca-Cola Enterprises achieve greater agility, cost savings, and global consistency, while ensuring compliance with local data sovereignty regulations