Cloud

https://landscape.cncf.io/

Characteristics

The NIST Definition of Cloud Computing (publication 800-145) is defined what is Cloud.

5 Essential Characteristics of the Cloud:

1. On-demand self-service
   • “a consumer can unilaterally provision computing capabilities… …as needed automatically without requiring human interaction with each service provider.
2. Broad network access
3. Resource pooling
   • “…resources are pooled to serve multiple consumers using a multi-tenant model…”
   • “There is a sense of location indepedence in that the customer generally has no control or knowledge over the exact location of the provided resources…”
4. Rapid elasticity
   • “Capabilities can be elastically provisioned and released…to scale rapidly outward and inward commensurate with demand”
   • “To the consumer, the capabilities available for provisioning often appear to be unlimited…”
5. Measured service
   • “Resource usage can be monitored, controled, and reported providing transparency for both the provider and consumer of the utilized service.”

Deployment Models

• Public
  • Azure, AWS, Google Cloud
  • Is easy
• Private
  • Be very skeptical
• Hybryd
  • Public + Public
  • Public + Private
  • Be skeptical
• Community
  • It exists

Service Models

Service model defines what end-users gets, manages, sees.

Unit of consumption (UoC) term is defined what you as a customer consume from service provider. Billing (cost) for service is based on UoC.

Three service models:

• IaaS - Infrastructure as a Service
  • UoC: VM or OS
  • Billing: Per second/minute/hour when VM is active
• PaaS - Platform as a Service
  • UoC: Runtime Environment (Node.js, PHP, etc.)
• SaaS - Software as a Service

IaaS -> PaaS -> SaaS:

• You loose flexibility
• You gain in terms of easy of management: less costs, less risks

AWS and Azure service models priorities and path.

• Azure more SaaS and PaaS
• AWS more IaaS and PaaS

Additional notable service models:

• FaaS - Function as a Service - when you upload a single function (serverless architecture). Examples: Azure Functions, AWS Lambda
• DBaaS - Database as a Service. Examples: Azure Cosmos DB, AWS DynamoDB, Google Spanner. They have fairly deep into performance tuning and configuration.