Computing Platforms & Technologies

Multiple cloud platforms have emerged to serve different organizational needs — from hyperscale public clouds to open-source private cloud frameworks. Understanding each platform's strengths, services, and positioning helps architects make informed provider decisions.

AWS (Amazon Web Services)

The market leader with the broadest and deepest service catalog (200+ services across 30+ geographic regions).

Core Services:

• Compute: EC2 (VMs), Lambda (serverless), ECS/EKS (containers)
• Storage: S3 (object), EBS (block), EFS (file), Glacier (archive)
• Database: RDS (relational), DynamoDB (NoSQL), Redshift (data warehouse), Aurora (managed MySQL/PostgreSQL)
• Networking: VPC, Route 53 (DNS), CloudFront (CDN), Direct Connect
• AI/ML: SageMaker, Rekognition, Comprehend, Bedrock (foundation models)

Strengths: Widest service breadth, largest ecosystem of partners and third-party tools, most mature IAM and compliance certifications (SOC2, HIPAA, FedRAMP)

Microsoft Azure

The second-largest cloud provider, with deep enterprise and Microsoft product integration.

Core Services:

• Compute: Azure VMs, Azure Functions (serverless), AKS (Kubernetes)
• Storage: Blob Storage, Azure Files, Azure Disk
• Database: Azure SQL Database, Cosmos DB (globally distributed NoSQL), Azure Synapse Analytics
• AI: Azure OpenAI Service, Cognitive Services, Azure ML
• Enterprise Integration: Active Directory, Teams, Office 365, Dynamics 365

Strengths: Hybrid cloud with Azure Arc and Azure Stack, dominant in enterprises running Windows Server / SQL Server workloads, best-in-class identity with Azure AD

GCP (Google Cloud Platform)

Google's cloud offering, known for data analytics, AI/ML, and Kubernetes (invented at Google).

Core Services:

• Compute: Compute Engine, Cloud Run (serverless containers), GKE (Kubernetes)
• Storage: Cloud Storage (object), Persistent Disk, Filestore
• Data & Analytics: BigQuery (serverless data warehouse), Dataflow (stream/batch), Pub/Sub (messaging)
• AI/ML: Vertex AI, AutoML, TensorFlow TPU hardware

Strengths: BigQuery for analytics at scale, superior networking (private global fiber backbone), leading AI/ML tooling (Gemini, Vertex AI), Kubernetes expertise

OpenStack

An open-source cloud operating system that powers private and community clouds. Not a public cloud — organizations deploy and operate it themselves.

Core Components:

• Nova: Compute (manages VMs, analogous to EC2)
• Neutron: Networking (SDN-based virtual networks)
• Cinder: Block storage (analogous to EBS)
• Swift: Object storage (analogous to S3)
• Keystone: Identity and authentication
• Horizon: Web-based dashboard

Deployments: CERN (particle physics), Walmart, Deutsche Telekom, many telecom carriers (ETSI NFV)

Strengths: No vendor lock-in, full source code access, customizable for specific hardware, GDPR-compliant data residency

Platform Landscape

Platform Selection Criteria

Multi-Cloud Reality

Most enterprises use 2+ providers: AWS for primary workloads, Azure for Microsoft integration, GCP for analytics. Tools like Terraform, Kubernetes, and Pulumi enable workload portability across platforms.