PLATFORM APPROACH

A platform can be deployed on-premises or in the cloud. In either case, an application platform typically comprises of three parts:

1. Hardware & System Software: An application needs an operating system and system software for it to run on any system hardware.
2. Infrastructure services: In a distributed environment, applications have to integrate with other services which are specialized like network services, storages services, identity services, infosec services etc.
   Points a) and b) are provided as a composite offering as Infrastructure as a Service (IaaS) by the Cloud majors like GCP, AWS, Azure etc. In case someone does not wish to use Cloud, they will have to create a similar environment on-premise. There are variations called Hybrid Cloud where part of the compute and services are on-premise and rest part is on the Cloud.
3. Application services: In a cloud platform approach, Applications run on top of IaaS. These are typically accessed by a Web Browser across devices like desktops, laptops, and tablets or via Mobile Apps by end-users. Applications are no longer monolithic but follow a service-oriented architecture which means applications access functions from other applications while making their functions accessible to other applications. The primary function of a set of applications is to provide services to end users and when deployed as above in “platform” approach, it is called an “application platform”.

We have to also discuss the Cloud in the same breath as Internet-based Application Platform could not have been possible without advances in Cloud Platforms which serves as the underlying IaaS.

Definition of Cloud

Demystifying Cloud Computing

Virtualization is the technology which enables Cloud computing.

A physical computer can be divided into multiple virtual machines each with its number of cores, memory and storage. All files and applications on one virtual machine cannot be accessed from another.

Virtual machines make more efficient use of the physical hardware hosting them. By running many virtual machines simultaneously, one physical server can become many virtual servers. A data center can thereby provide flexible compute to a number of organizations. Such an approach lowers the cost for everyone.

Users can access cloud services in a device-agnostic manner using a simple browser or App via Internet.

Types of Cloud Computing

Software-as-a-Service (SaaS): Applications are hosted on the cloud servers and accessed over Internet without requiring a local installation of the same. You need not own the Application to use it. It is available “on tap” and typically follows the “pay-per-use” model or a monthly subscription. Examples of SaaS applications include Salesforce, Office 365 etc.

Platform-as-a-Service (PaaS): Instead of Application the Platform operates like in a rented model which consists of the infrastructure. Users can build and run their own applications on top of it. PaaS vendors offer development tools, infrastructure, and operating systems with all resources accessible via the Internet. PaaS examples include Microsoft Azure.

Infrastructure-as-a-Service (IaaS): Under this model, a company can purchase on “pay-as-you-use” model all the servers and storage need from a Cloud Service Provider. Thereafter, they can build and run their applications on this infrastructure. Example of IaaS providers are DigitalOcean, Google Compute Engine etc.

Function-as-a-Service (FaaS): This is a serverless computing model. Cloud applications are broken down into even smaller components that only run when they're needed. This is an emerging trend and can help control the spend on the Cloud. FaaS or serverless applications still run on Cloud servers but not on dedicated machines. The companies building such applications do not have to manage any specific servers.

Cloud Deployment Paradigms

The below are the different cloud deployment types based on location of the cloud servers are and based on how they are managed:

• Private cloud: A private cloud can be a single server, a whole data center, or an entire distributed network dedicated to an organization.
• Public cloud: A public cloud as the name suggests is used by multiple tenants. It is a service run by a cloud service provider that include multiple virtualized servers in one or multiple data centers.
• Hybrid cloud: Hybrid cloud deployments are a combination of public and private clouds, and may even include servers which are deployed on-premises.
• Multicloud: Multicloud is a type of cloud deployment that involves using multiple public clouds.

Microservices – An upcoming Paradigm

As per definition in www.microservices.io, Microservices or Microservice architecture – application is structured as a collection of services that are

• Highly maintainable and testable
• Loosely coupled
• Independently deployable
• Organized around business capabilities
• Owned by a small team
• The microservice architecture enables the rapid, frequent and reliable delivery of large, complex applications.
• This is also a structured way to evolve the technology stack

Each service is flexible, robust, composable and complete. They run as autonomous processes and communicate with one another through APIs

What are containers?

The container platform runs on top of the host operating system. Containers run the binaries, libraries along with the application. Unlike Virtual Machines, containers do not contain a guest operating system which ensures that containers are kept light.