Unit 4: Cloud platforms in industry
Cloud Computing notes · PTU syllabus (PGCA1937)
On this page
Unit summary
The major providers turn cloud concepts into services. This unit covers Amazon Web Services compute, storage and communication services, Google App Engine's architecture and application life cycle, Microsoft Azure core concepts, SQL Azure and the Windows Azure Platform Appliance.
After this unit you can
- Describe AWS compute, storage and communication services
- Explain the architecture and life cycle of Google App Engine applications
- Explain Microsoft Azure core concepts
- Describe SQL Azure and the Azure platform appliance
PTU syllabus topics
- Amazon Web Services compute/storage/communication services
- Google App Engine architecture and application lifecycle
- Microsoft Azure core concepts
- SQL Azure
- Windows Azure Platform Appliance
Topic 1
Amazon Web Services: compute services
- Amazon EC2
- Virtual servers (instances) of many types; on-demand, reserved and spot pricing; AMIs as templates
- Auto Scaling and Elastic Load Balancing
- Adjust instance counts; distribute traffic
- AWS Lambda
- Serverless functions run on events, billed per request and duration
- Containers
- ECS, EKS (Kubernetes) and Fargate
- Elastic Beanstalk
- PaaS-style deployment of web applications
Example
Choosing AWS compute: a website runs on EC2 instances behind a load balancer; an image-resize function runs on AWS Lambda only when a file is uploaded, charging per execution.
Exam tip
EC2 = virtual servers (IaaS); Lambda = serverless functions; Elastic Beanstalk = PaaS-style deployment.
Topic 2
AWS storage services
- Amazon S3
- Object storage in buckets; storage classes from Standard to Glacier
- Amazon EBS
- Block volumes for EC2
- Amazon EFS
- Elastic shared file system
- Databases
- RDS (relational), DynamoDB (NoSQL), Redshift (warehouse)
- Content delivery
- CloudFront CDN
Example
Storage classes: a company keeps frequently used images in S3 Standard, older logs in S3 Infrequent Access, and seven-year archives in S3 Glacier at much lower cost.
Exam tip
S3 = object storage; EBS = block storage for EC2; EFS = shared file storage.
Topic 3
AWS communication services
- Amazon SQS
- Message queues decoupling components
- Amazon SNS
- Publish–subscribe notifications (email, SMS, HTTP)
- Amazon EventBridge
- Event bus routing events between services
- Amazon SES
- Email sending
- Networking
- VPC (private networks), Route 53 (DNS), Direct Connect
Example
A results website: users hit CloudFront and a load balancer, EC2 instances in an Auto Scaling group serve pages, marks are stored in RDS, PDFs in S3, and SQS queues e-mail jobs that SES sends.
Topic 4
Google App Engine: architecture
- Google App Engine (GAE): a fully managed PaaS for web applications and APIs; developers upload code and Google handles servers, scaling and load balancing.
Environments
Standard (sandboxed runtimes, scale to zero) and flexible (Docker containers on VMs)
Front ends and load balancers
Route requests to instances
Application instances
Auto-scaled by traffic
Services
Datastore or Firestore, Cloud SQL, Memorystore cache, Cloud Tasks, Cron jobs
Management
Versions, traffic splitting, logging and monitoring
Example
Deploying on App Engine: a developer writes a Python web app and runs gcloud app deploy. App Engine provisions servers, balances load and scales instances automatically with traffic.
Exam tip
App Engine is PaaS — developers manage code, not servers.
Topic 5
Application life cycle on App Engine
- 1
Develop locally with the SDK or gcloud CLI
- 2
Configure app.yaml (runtime, scaling, handlers)
- 3
Deploy with gcloud app deploy — a new version is created
- 4
Test the version URL; split or migrate traffic
- 5
Monitor logs, errors and metrics
- 6
Update with new versions; delete old ones
Example
From code to production: develop and test locally, deploy a new version with gcloud, split 10% of traffic to the new version to test it, then move all traffic once it is stable, and monitor logs.
Exam tip
Versions and traffic splitting allow safe updates.
Topic 6
Microsoft Azure: core concepts
- Regions and availability zones
- Geographic locations and isolated data centres
- Subscriptions and resource groups
- Billing boundary; logical containers for related resources
- Azure Resource Manager
- Deploys and manages resources, with ARM or Bicep templates
- Compute
- Virtual Machines, App Service, Azure Functions, AKS
- Storage
- Blob, Files, Queues, Tables, Disks
- Identity
- Microsoft Entra ID (formerly Azure Active Directory)
- Azure grew from Windows Azure (2010), whose original model ran web and worker roles in Cloud Services, with a fabric controller managing machines.
Example
Organising Azure: a company groups a website's VM, database and storage into one resource group in the Central India region, so they can be managed, monitored and deleted together.
Exam tip
Key concepts: regions, availability zones, subscriptions and resource groups.
Topic 7
SQL Azure
- SQL Azure (now Azure SQL Database) is Microsoft SQL Server delivered as a managed relational database service: automatic backups, patching, high availability and scaling, with familiar T-SQL and tools.
Azure SQL Database (single or elastic pool)
Fully managed database; serverless option
New cloud applications
Azure SQL Managed Instance
Near-complete SQL Server compatibility
Lift-and-shift of existing databases
SQL Server on Azure VMs
Full control of the server
Legacy features or special configurations
Example
A managed database: a company moves its SQL Server database to Azure SQL Database. Backups, patching and high availability are handled by Microsoft, and the company scales performance tiers as needed.
Exam tip
Azure SQL Database is DBaaS — no server or OS to manage.
Topic 8
Windows Azure Platform Appliance
- Announced in 2010, the Windows Azure Platform Appliance was to let large organisations and partners run the Azure platform (compute, storage, SQL Azure) on Microsoft-specified hardware in their own data centres — a private or hosted cloud consistent with the public Azure.
- It never shipped widely; the idea returned as Azure Stack (2017) and is now Azure Stack Hub and Azure Local, which bring Azure services to on-premises hardware for hybrid cloud, sovereignty and disconnected scenarios.
Example
What it aimed to do: large organisations and hosting partners were meant to run Azure's platform inside their own data centres. The idea later evolved into today's Azure Stack (Azure Stack Hub) for hybrid cloud.
Exam tip
Link the appliance to the modern Azure Stack hybrid offering.
Key terms
- EC2
- AWS virtual server service
- S3
- AWS object storage service
- PaaS
- Platform where the provider manages servers and runtime
- Resource group
- Container for related Azure resources
- Azure SQL Database
- Managed relational database service from Microsoft
Quick revision
- EC2, Auto Scaling, ELB, Lambda, containers, Beanstalk.
- S3, EBS, EFS, RDS, DynamoDB, Redshift, CloudFront.
- SQS, SNS, EventBridge, SES, VPC, Route 53.
- App Engine standard and flexible; app.yaml; versions and traffic splitting.
- Azure regions, subscriptions, resource groups, ARM, compute, storage, Entra ID; SQL Azure options; platform appliance → Azure Stack.
Important exam questions
Practice questions written to the PTU exam pattern for this unit's syllabus: short answers (Section A style) and long answers (Sections B and C style).
Short-answer questions
- Q1.What is Amazon EC2?
- Q2.Distinguish SQS and SNS.
- Q3.What is AWS Lambda?
- Q4.What does app.yaml configure in App Engine?
- Q5.What is an Azure resource group?
- Q6.What became of the Windows Azure Platform Appliance idea?
Long-answer questions
- Q1.Explain AWS compute, storage and communication services.
- Q2.Explain the architecture of Google App Engine and the application life cycle.
- Q3.Explain the core concepts of Microsoft Azure.
- Q4.Explain SQL Azure and the Windows Azure Platform Appliance.
Stuck on this unit?
Message SBS on WhatsApp for help with Cloud Computing, or to ask about studying M.Sc IT at Synetic.
