Module 1 - Cloud Concepts

Why Cloud Services?

Cloud Computing

• Cloud computing = Delivery (wide ranges) of computing services over the internet (the cloud).
• Cloud service provider = The company that delivering such services

Key Concepts

• High Availability = Ability top keep service up and running for a long period of time
• Scalability = Ability to increase/decrease resources for any given workload
  • Scale out = add additional resource
  • Scale up = add additional capability to existing resources
• Elasticity = Automatic and dynamic scalability
• Agility = Ability to react quickly
  • allocate/deallocate resource on-demand via self-service w/o manual intervention
• Fault Tolerance = Ability to remain up & running  even a component or service no longer functioning 
  • archived by 'redundancy'
• Disaster Recovery = Ability to recover from an event which has taken  down a cloud service
• Global Reach = Ability to reach audiences in various regions around the globe 
• Customer latency capability
  • Customers are experiencing slowness = customer are experiencing some latency 
  • Could caused by the service is not local to the customer
  • Could be solved by deploying services in data centers around the globe
• Predictive cost consideration = Ability for user to predict cost incurs for cloud services they use
• Technical skill consideration
  • User who expert in application they want to run without the need of skills to build & maintain underlying h/w & s/w infrastructure
• Increased Productivity
  • Data center requires lots of h/w setup (called 'racking & stacking'), s/w patching
  • Cloud computing eliminates these tasks so IT department can focus on important business goals
• Security
  • Cloud providers usually provide broad set of policies, technology, controls more than any single company can has on their own
• See cloud computing terms for more

Economy of Scale

• Ability to reduce cost and gain efficiency when operating at larger scale  vs. smaller scale
• Cloud provider can acquire h/w at much lower cost and pass on that benefits to customers

CapEx vs. OpEx

• Capital Expenditure (CapEx) = spending money on physical infrastructure up front and values reduce over time
• Operational Expenditure (OpEx) = spending money on services or products now and being billed for them now
• Consumption-based Model
  • Pay only what you use
  • No up-front cost/CapEx
  • Can pay more to get additional resource when needed and can stop paying when no longer need it

Types of Cloud Models

Public Cloud vs. Private Cloud

	Public Cloud	Private Cloud
Data center	Cloud provider owned	Organization-owned
Ownership	Cloud service provider	The organization itself
Users	Multiple organizations	Within the organization
Access	Public	Single organization
Connectivity	Over the internet	Usually over private network
Skill to use	No deep tech skill required	Require deep tech skills to setup, manage, and maintain

Hybrid Cloud

• Combined both Public and Private cloud
• Some resources run on public cloud, some on private cloud
• Organization retain the management control in private cloud
• Use cases
  • Highly secured data or something that cannot put on public cloud e.g. due to legal reasons
  • Old specific hardware that can't be updated e.g. mainframe

Comparisons

Public Cloud	Private Cloud	Hybrid Cloud
Pros No CapEx Agility Consumption-based No maintenance No deep skill required	Pros Complete control over resources and security Compliance achievable Specific scenario achievable	Pros Flexibility - most flexible Costs - still can take advantage from public cloud Control & Security - can still put in private cloud Compliance still achievable Specific scenario achievable
Cons Security requirement may not be met Compliance may not be met Ownership - cannot manage as they wish Specific scenario - legacy app or h/w	Cons Up-front CapEx Agility Maintenance Skills	Cons Up-front CapEx Costs Skills - including make both operate together Ease of management - should have guidelines on what on public or private cloud

Shared Responsibility Model

• Public cloud advantages do not absolve the customer from protecting their users, applications, and service offering.
• Shared responsibility model ensures cloud workloads are run securely and in a well-managed way.
  • Cloud provider is responsible for some aspect of the workload management.
  • Customer or end user is also responsible for other aspects
  • Both share a responsibility

IaaS	PaaS	SaaS
Characteristics Most basic but most flexible Rent only IT infrastructure No upfront cost	Characteristics Provide environment for building, testing, and deploying s/w Without worrying about underlying infrastructure No upfront cost	Characteristics Software that centrally hosted and managed e.g. Office 365 Usually licensed through monthly or annual subscription No upfront cost
Common usage scenarios Migrating workloads - easy to migrate from on-premise Test and development environment - quicker to deploy Website hosting - lower cost Storage, backup, and recovery - more simplified	Common usage scenarios Development framework - developer can build upon Analytics or Business Intelligence - analyze and mine data	Common usage scenarios Office 365 Skype Microsoft Dynamics CRM Online

Comparisons

IaaS	PaaS	SaaS
Pros No CapEx Agility Consumption-based Skills - no deep technical to deploy & use Cloud benefits - secured and HA Flexibility - control to configure, manages h/w	Pros No CapEx Agility Consumption-based Skills - no deep technical to deploy & use Cloud benefits - secured and HA Productivity - focus on application development	Pros No CapEx Agility Pay-as-you-go - monthly or yearly Flexibility - data accessible from anywhere
Cons user manages and maintains services they provision	Cons Some limitations from PaaS	Cons Software limitations - no or limited customizations