Cloud Computing Platforms and Infrastructure Cloud Computing Spring 2025

2.1 Overview of Major Cloud Service Providers

• Cloud computing has revolutionized the way organizations
deploy, manage, and scale their IT infrastructure.
• The three dominant players in the cloud computing market are
Amazon Web Services (AWS), Microsoft Azure, and Google
Cloud Platform (GCP).
• Each of these providers offers a comprehensive suite of
services, but they differ in their strengths, pricing models, and
target audiences.
Amazon Web Services (AWS)
AWS, launched in 2006, is the most mature and widely adopted cloud
platform. It offers over 200 services, including computing power (EC2),
storage (S3), databases (RDS), machine learning (SageMaker), and more.
Strengths: AWS is known for its extensive global infrastructure,
scalability, and a vast ecosystem of third-party integrations. It is
particularly popular among startups and enterprises for its flexibility and
pay-as-you-go pricing model.
Use Cases: AWS is ideal for businesses looking for a broad range of
services, from web hosting to big data analytics and IoT.
 Amazon Web Services (AWS)
AWS offers a broad range of services, including:
• Compute Services: EC2 (Elastic Compute Cloud), Lambda (serverless computing)
• Storage Services: S3 (Simple Storage Service), EBS (Elastic Block Store)
• Networking: VPC (Virtual Private Cloud), Route 53 (DNS service)
• Database Services: RDS (Relational Database Service), DynamoDB (NoSQL
database)
• AI/ML: SageMaker, Rekognition
• SageMaker is a managed service that automates the building and deploying of
machine learning data models.
• Rekognition automates image recognition and video analysis for your applications
without machine learning (ML) experience.
Microsoft Azure
Azure, launched in 2010, is Microsoft's cloud platform. It provides a
wide array of services, including virtual machines, AI and machine
learning (Azure ML), and enterprise applications like Office 365 and
Dynamics 365.
Strengths: Azure is deeply integrated with Microsoft’s software
ecosystem, making it a natural choice for enterprises already using
Windows Server, Active Directory, or other Microsoft products. It also
offers strong hybrid cloud capabilities, allowing seamless integration
between on-premises and cloud environments.
Use Cases: Azure is well-suited for enterprises with existing Microsoft
Microsoft Azure
Key Azure services include:
Compute: Virtual Machines, Azure Functions (serverless computing)
Storage: Blob Storage, Azure Files
Networking: Virtual Network, Azure Load Balancer
Database: Azure SQL Database, Cosmos DB
AI & Analytics: Azure Machine Learning, Power BI integration
Google Cloud Platform (GCP)
GCP, launched in 2011, is Google’s cloud offering. It is known
for its strengths in data analytics, machine learning
(TensorFlow), and container orchestration (Kubernetes).
Strengths: GCP excels in big data and machine learning,
leveraging Google’s expertise in these areas. It also offers
competitive pricing and strong performance for data-intensive
applications.
Use Cases: GCP is ideal for organizations focused on data
analytics, AI/ML, and containerized applications.
Containerized Applications
• Containerized applications are applications run in isolated packages of code called
containers.
• Containers include all the dependencies that an application might need to run on
any host operating system, such as libraries, binaries, configuration files, and
frameworks, into a single lightweight executable.
• Software developers use containerization to deploy applications in multiple
environments without rewriting the program code.
• They build an application once and deploy it on multiple operating systems.
• For example, they run the same containers on Linux and Windows operating
systems.
Google Cloud Platform (GCP)
Key services include:
Compute: Compute Engine, Cloud Functions
Storage: Cloud Storage, Persistent Disks
Networking: Cloud Load Balancing, VPC
Database: BigQuery, Cloud Spanner
AI & ML: TensorFlow, AutoML
2.2 Cloud Architecture and Data Centers
• Cloud architecture refers to the design and structure of cloud
environments, including the components and subcomponents
required for cloud computing.
• At the heart of cloud architecture are data centers, which house
the physical infrastructure (servers, storage, networking
equipment) that powers cloud services.
Cloud Architecture Layers
Cloud computing infrastructure is built on a network of global data centers that
provide scalable and high-availability services. Cloud architecture typically consists
of the following layers:
 Infrastructure Layer: Physical hardware, data centers, networking components
 Virtualization Layer: Hypervisors, virtual machines (VMs), containers
 Platform Layer: Middleware, APIs, and orchestration tools
 Application Layer: SaaS applications, cloud-native services
Key Components of Cloud Architecture
1. Front-End: The client-side interface that users interact with
(e.g., web browsers, mobile apps).
2. Back-End: The cloud infrastructure, including servers,
storage, and databases.
3. Network: The communication channels that connect front-
end and back-end components.
4. Middleware: Software that enables communication and data
management between applications.
Data Centers and Regions
Cloud providers operate data centers globally, categorized into:
 Regions: Geographically distinct areas with multiple data centers
 Availability Zones (AZs): Multiple data centers within a region, ensuring
redundancy
 Edge Locations: Content delivery and caching points for faster access
Data Centers
Design: Modern data centers are designed for high availability,
scalability, and energy efficiency. They are often distributed across
multiple geographic regions to ensure redundancy and low latency.
Global Infrastructure: Major cloud providers operate data centers in
multiple regions and availability zones. For example, AWS has regions in
North America, Europe, Asia, and more, each consisting of multiple
isolated data centers.
Sustainability: Cloud providers are increasingly focusing on renewable
energy and energy-efficient designs to reduce the environmental impact
of data centers.
2.3 Resource Management in the Cloud
• Resource management in the cloud involves allocating and
optimizing computing resources such as CPU, memory, storage,
and network bandwidth to meet application demands efficiently.
Resource Management in the Cloud
Cloud providers enable efficient resource management through:
 Elasticity: The ability to scale resources up or down as demand
fluctuates
 Multi-tenancy: Shared infrastructure among multiple users while
ensuring security and isolation
 Monitoring & Optimization: Tools like AWS CloudWatch, Azure
Monitor, and Google Stackdriver help track resource usage and
performance
Key Aspects of Resource Management
Provisioning: Allocating resources to applications or users based
on demand.
Monitoring: Tracking resource usage to identify bottlenecks or
underutilization.
Optimization: Adjusting resource allocation to improve
performance and reduce costs.
Cost Management: Using tools like AWS Cost Explorer or
Azure Cost Management to monitor and control cloud spending.
Challenges of Resource Management
Over-Provisioning: Allocating more resources than necessary,
leading to higher costs.
Under-Provisioning: Allocating insufficient resources, resulting
in poor performance.
Dynamic Workloads: Managing resources for applications with
fluctuating demand.
Virtual Machines vs. Containers
 Virtual Machines (VMs): Provide full OS-level virtualization, enabling isolated
environments
 Containers: Lightweight, portable environments (e.g., Docker, Kubernetes) with
faster deployment and scaling
2.4 Load Balancing and Auto-Scaling
Load Balancing:
Load balancing distributes incoming network traffic across
multiple servers to ensure no single server is overwhelmed. This
improves application availability, reliability, and performance.
Cloud-based load balancing options include:
 AWS Elastic Load Balancer (ELB)
 Azure Load Balancer
 GCP Cloud Load Balancing
Types of Load Balancers
Application Load Balancer (ALB): Operates at the application
layer (Layer 7) and is ideal for HTTP/HTTPS traffic.

Network Load Balancer (NLB): Operates at the transport layer

(Layer 4) and is suitable for TCP/UDP traffic.

Global Load Balancer: Distributes traffic across multiple

regions for global applications.
Auto-Scaling
Auto-scaling automatically adjusts the number of compute
resources based on real-time demand. This ensures optimal
performance during peak times and cost savings during low
traffic.
Vertical Scaling: Increasing the capacity of existing resources
(e.g., adding more CPU or memory).
Horizontal Scaling: Adding more instances of a resource (e.g.,
launching additional servers).
Auto-Scaling
Major cloud auto-scaling services include:
 AWS Auto Scaling
 Azure Scale Sets
 GCP Managed Instance Groups
By leveraging these cloud infrastructure components, organizations can
achieve high availability, scalability, and cost efficiency in their IT
operations.
2.5 Cloud Networking: VPCs, Subnets, and Firewalls
Cloud networking involves the configuration and management of network resources in the
cloud.
Virtual Private Cloud (VPC): A VPC is a logically isolated section of the cloud where you
can launch resources in a virtual network.
Subnets: Subdivisions of a VPC that allow you to segment resources for security and
performance.
Firewalls: Security groups and network access control lists (ACLs) that control inbound and
outbound traffic to resources.
Key Features:
Private and Public Subnets: Public subnets allow internet access, while private subnets are
isolated for sensitive resources.
VPN and Direct Connect: Secure connections between on-premises networks and the cloud.
Cloud Service Orchestration and Automation
Tools
Automation is crucial for managing cloud infrastructure efficiently.
Popular tools include:
• Terraform
• Kubernetes
• Edge Computing
• Fog Computing
• Cloud Formation & ARM Templates
Terraform
Terraform is an infrastructure-as-code (IaC) tool that enables you
to define and provision cloud resources using declarative
configuration files.
Benefits: Version control, repeatability, and consistency in
resource provisioning.
Use Cases: Automating the deployment of complex cloud
environments.
Declarative configuration management
Declarative configuration management refers to the class of tools
that allow operators to declare a desired state of some system (be
it a physical machine, an EC2 VPC, an entire Google Cloud
account, or anything else), and then allow the system to
automatically compare that desired state to the present state, and
then automatically update the managed system to match the
declared state.
Kubernetes
Kubernetes is an open-source container orchestration platform
that automates the deployment, scaling, and management of
containerized applications.
Key Features:
Pods: The smallest deployable units in Kubernetes.
Services: Enable communication between pods.
Scaling: Automatically adjusts the number of pods based on
demand.
Edge Computing
Edge computing brings computation and data storage closer to
the devices where data is generated, reducing latency and
bandwidth usage.
Processing data at or near the source (e.g., IoT devices,
autonomous vehicles) instead of centralized cloud data centers.
Use Cases: IoT, real-time analytics, and autonomous vehicles.
Examples: AWS IoT Greengrass, Azure IoT Edge.
Fog Computing
Fog Computing:
Fog computing extends cloud computing to the edge of the
network, enabling data processing at intermediate points between
the cloud and end devices.
Distributes computing resources between edge devices and the
cloud to improve efficiency and scalability.
Use Cases: Smart cities, industrial automation.
Examples: Cisco IOx, OpenFog Consortium.
Conclusion
Cloud computing platforms and infrastructure form the backbone
of modern IT systems.
Understanding the offerings of major cloud providers, the
architecture of data centers, and the tools for resource
management, networking, and automation is essential for
designing and managing scalable, efficient, and secure cloud
environments. As cloud technologies continue to evolve,
concepts like edge and fog computing are becoming increasingly
important, enabling new possibilities for real-time data
processing and IoT applications.