Overview

Cloud Computing has matured over time to offer a platform for businesses to scale on-demand with increased capacity to provision compute infrastructures on-the-fly. This agility has created opportunities that businesses and enterprises can leverage as a strategy to optimize scarce and costly resources on a pay-as-you-go basis with immediate benefits. Fig 1: Uses of Cloud Computing

Overarching Benefits:

Pay-As-You-Go pricing (cost management)
Elasticity (ability to scale)
Agility (speed to market)
Flexibility (deployment options)
Resource Pooling

Every Cloud Strategy is premised on a composition of deployment models that suits a business, its short term goals and long term vision. The alignment of the cloud strategy to the vision of the business creates a perfect blend of opportunities for growth, value and customer/user satisfaction.

It Starts With The Business Architecture

This provides a map of the business intent and the critical components driving the cloud strategy for adoption. Fig 2: Business Architecture as a tool for Strategy Execution

The Key components include:

Business Capabilities: This involves analysing the current and future state of the business and mapping its KPIs to a cloud strategy for adoption
Value Streams: This is the assessment of the current value chains in the business using SWOT (strength, weakness, opportunities & threats) models
Gap Analysis of Processes: This is the evaluation of the IT workflow gaps in the business and to prioritize technology solutions
Risks Assessment: Business risks are identified and prioritized for mitigation. Risk-aware cloud strategies adopted
Strategic Action Plans: This is the creation, adoption and implementation of the cloud deployment road-map and recommended framework for delivery
Fig 3: Business Architecture as a tool for Digital Transformation

Once the business architecture is laid out, next is to:

Identify the Business and Technical requirements (The What): This describes how the application, system or service functions.
What Problems are We trying to Solve (The Why): This evaluates the business drivers for leveraging the Cloud and its overall benefits.
Understand the Change Management and Deployment Challenges (The How): This assesses the organization's technology posture and readiness to adopt the Cloud strategy.

Cloud Service Models and Features

The concept of resource pooling of IT infrastructure (data-centre and network) creates unique opportunities for deploying the cloud. These include:

Access to a broad network backbone
Ability to scale
Pay for what You Use
On-demand provisioning of resources

The Multi-Cloud Service models provides more agility for businesses to focus on scaling their operations and less time on managing infrastructure. The usual three reference models open to businesses to compare their Cloud adoption strategies are:

Software as a Service (SaaS)
Platform as a Service (PaaS)
Infrastructure as a Service (IaaS) Fig 4: Features of the Cloud Service Models

SaaS: Software as a Service: This cloud application service focuses on access to on-demand ready to use Cloud Applications as a software solution package that can be deployed in the cloud market. It is dependent on a web delivery model and eliminates the cost element from businesses in managing the overheads of committing internal IT staff, potential technical issues and maintenance.

Use case for SaaS Cloud Package: The underlying indicators of when to use software as a service strategy:

Ideal for start-ups or small businesses building a cloud solution entirely from the ground up
Eliminating the need for capital expenses or high operating costs
When speed to deliver short-term projects is critical
When the business doesn’t have the technical expertise to build a management software to handle orchestration of applications and its dependencies
Easy access to SaaS applications via web based internet access from client devices or program interfaces.

Restrictions of SaaS model: The limitations include:

No control over underlying cloud management infrastructure and dependencies (servers, network, storage, data security and operating systems).
Custom features are limited and basic
Licenses lock-in users for this cloud solution with limited integration to use proprietary software
Performance and availability are not guaranteed but managed with service level agreements (SLAs)

SaaS Examples:

Xero (accounting)
CRM (Salesforce)
ERP (SAP)
Email (Microsoft 360)
Storage (Dropbox)

PaaS: Platform as a Service: This Cloud platform services focuses on providing a platform to run components of software applications in a virtualised environment as a platform solution package that can be deployed in the cloud market. It is dependent on a framework of several managed compute resources (servers, storage and network) for developers as a base to create customised applications. This model also eliminates the cost element at the backend while the overlaying critical business applications are created, run and managed by the developer team.

Use case for PaaS Cloud Package: The underlying indicators of when to use platform as a service strategy includes:

Ideal for small and established businesses with known and predictive compute capacity
Provides a low cost compute environment to create, test and deploy software applications
When the business has the technical expertise to build and manage software applications on-prem and in a virtualised environment
Easy integration of software applications to web services and databases to a hybrid multicloud model
To automate business applications
To create scalable and highly available middleware platform to run software applications

Restrictions of PaaS model: The limitations include:

Data security options are limited in the testing and deployment of cloud applications due to hosting policies.
Data archival and backup options are limited and less flexible.
Licenses lock-in users for this cloud solution with limited integration to use proprietary software.
Performance and availability are not guaranteed but managed with service level agreements (SLAs).

PaaS Examples:

OpenShift
Google App Engine
Heroku
Force.com
AWS Elastic Beanstalk

IaaS: Infrastructure as a Service: This Cloud Infrastructure services focuses on providing highly scalable and automated virtualized network and data-centre compute resources as a Cloud Infrastructure solution package that can be deployed in the cloud market. It is a flexible delivery model that allows businesses to purchase right-sized compute resources on-demand on pay-as-you-go basis. This eliminates the time delays, high capital and operating expenses in provisioning critical compute infrastructure.

Use case for IaaS Cloud Package: Underlying indicators of when to use Infrastructure as a service strategy:

Ideal for start-ups, small and established businesses with high performance and scalability requirements for its three tier web architecture
Flexible and dynamic utility pricing of compute resources
Agility in provisioning compute resources on-demand
Provides redundancy and fail-over access across hybrid multicloud deployments
Offers businesses control over applications and infrastructure on-prem or virtual data-centers

Restrictions of IaaS model: The limitations include:

Legacy applications and systems have limited performance and security controls
Virtualised compute resources are still vulnerable to security misconfigurations at the back-end including workloads
Vendor lock-in with the use of proprietary Infrastructure-as-a-service offerings as a cloud solution creates a point-of-failure
Performance and availability are not guaranteed but managed with service level agreements (SLAs)

IaaS Examples:

Google Compute Engine
DigitalOcean
Rackspace
Amazon Web Services (AWS)
Microsoft Azure

Strategies of Cloud Deployment Models

The cloud infrastructure is similar to the network and data-centre but virtualised and provisioned for the public to use. The common cloud deployment models:

Public Cloud: This is a type of Cloud Infrastructure or platform that supports the use of compute resources by multiple users or businesses in a shared environment on a subscription basis. It is mostly used for application development, testing and running non-mission critical workloads on a provider’s data-centre. Fig 5: Features of the Public Cloud
Private Cloud: This is also a type of Cloud Infrastructure or platform used by a single organization to run its IT workloads. Fig 6: Features of the Private Cloud
Hybrid Cloud: This type of cloud deployment supports the use of private and public cloud infrastructure based on standardized compute technologies that are portable.

Hybrid Cloud.JPG Fig 7: Features of the Hybrid Cloud

Community Cloud: This is a type of deployment that supports the pooling, sharing and access to compute resources by multiple organizations that are part of a group, business, conglomerate or community. Access is limited to only members of the Community. Fig 8: Features of the Community Cloud

Selecting the Right Cloud Strategy for the Business

It is essential to understand the underlying critical KPIs of the business and to align it to the overall vision using cloud technology solutions to lead and drive the transformation.

The key considerations when architecting the right cloud strategy are:

Business and Technical requirements: This focuses on the critical aspects of business continuity, performance, scalability, security, regulation and disaster recovery.
Financial (cost-benefit ratio): This focuses on the total cost of ownership in implementing the cloud strategy and the benefits.
Strategic Opportunities: This focuses on distinct advantages of the cloud adoption initiative (eg. Speed-to-market or right-sizing of compute resources), maximizing and allocating available surplus capacity to value streams.
Organization Structure and Cloud Maturity: This focuses on the business capabilities of the organization and its IT maturity in deploying cloud solutions.
Risk Appetite: This focuses on the risk levels a business is ready to tolerate in implementing a selected cloud strategy. A risk mitigation plan is adopted to manage the impact.

Recommendation:

Choosing the right cloud service model and deployment solution should be based on the:

Critical drivers of the business requirements
Change management constraints in implementing the Cloud Strategy
Customer and Business risks

The selection process of the best-for-purpose cloud solution should include a map out of the Cloud Service Models to deployment strategies based on the existing and future "business architecture". Cloud Service_and_Deployment.JPG Fig 9: The Cloud Strategy Matrix