Cloud Native: Evolution Story of Software Infrastructure

This article was originally posted on iauro's website here

On its journey to produce DevOps experts, iauro has got its hands around one of the most revolutionary infrastructures of our times - Cloud Native. Native Infrastructure has been attracting attention recently owing to the situation that the industry is forced into. We therefore decided to share our knowledge on the subject in detail and depth, with a series of articles. We deemed it right to start with talking more about what Infrastructure is and how we arrived at the cloud. This journey has a lot of insights about how administration has also evolved around this subject. Let us begin.

An Introduction to  Infrastructure

Every application needs an anatomy of software and hardware that can provide all the required assistance for its execution. This anatomy, called Software Infrastructure, typically consists of:

• Operating System
• Data Repository
• Configuration Management
• Deployment Pipeline
• Additional software hardware specific to the application needs

The modern day applications are evolving to Native Infrastructure. However, that wasn’t always the case. While we will talk more about cloud later in this article, let us first go to the very beginning of the infrastructure’s journey and see how it all started.

The Initial Space

In order to produce and run large scale software, organizations needed a lot of space. By space, we mean, actual physical rooms. These rooms had to be air-conditioned, sealed and very very skeptical of human intrusion. We’re talking about the times when infrastructure was provided by Servers. These big, noisy and very expensive electronic machines required a lot of power and people to keep them running. In return, here’s how they ‘served’

• As per the business requirements, they could be configured to provide storage space, operation environment, development resources etc.
• Servers are very powerful. With relatively low failure rates they can last a very long time 
• Servers are essentially the pioneers of applications infrastructure. They laid the basis for the future infrastructure technologies to thrive.

The physical existence of servers was their biggest disadvantage. They were so big that buying them was actually a capital expenditure for the organization. Apart from that there were a few functional disadvantages too. To name a few

• Running Multiple applications at large scale was difficult
• Network Routing was complicated as the number of applications and user increased
• Multiple users accessing the server also slows down its performance drastically.

The Virtual Solution

The next logical step from server was to attain their physical space in a virtual environment. This kind of infrastructure was, in fact named Virtualization. It was a helpful step because

• Virtual servers are configurable on software so the physical resource utilisation is reduced. 
• These Virtual Machines (VMs) are essentially emulators for physical servers, so they practically never wear out.
• One server can run multiple VMs
• Portability was introduced to infrastructures as VMs could be moved between different servers

Virtualization was definitely a step forward. The only problem was that the companies still needed to spend on servers and the resources to maintain them. The number of servers was reduced thanks to VMs but the necessity was still there. 

Infrastructure as a Service

Infrastructure as a Service (IaaS) was the primitive step that evolved into Native Infrastructure. 

IaaS was provided in the form of a networking, storage and computation environment which could be consumed on demand. Additional services like provisioning and inventory management were also handled by this infrastructure. Essentially, the organization could get rid of all the hardware and get its VMs from third parties on demand. The benefits were obvious:

• Less Physical space required along with reduced expenditure on maintenance resources.
• The service was accessed using Application Programming Interfaces (APIs). API calls were fast and could be easily manipulated as per requirement

This was a new idea and organizations were still not able to completely shift towards it. In fact, many companies used IaaS in a traditional way by creating space for their own servers and using them to create the services.

Platform as a Service

Platform as a Service (PaaS) were more of a complement to IaaS rather than a step ahead. While IaaS hid the physical servers, PaaS were meant to abstract the operating system. However, there was one thing that cleared the way for Native Infrastructure.

PaaS made application development platform independent. This meant that the developer were free to write their code without worrying about anything but the business logic that was expected from that code. This led the development lifecycle into a new generation. New business ideas emerged and new ventures were start up and small organizations were a little relieved from the overhead of physical space and other similar expenses. This was also the time when Public PaaS providers emerged in the market. PaaS had given birth to a new business vertical altogether

Moving to the Clouds

The stage was set. Infrastructure had become an operational expense rather than a capital one. Since, operational costs are billed by the time of consumption the aim was now to minimize this time. This lead to a revolutionary infrastructure technology - Native Infrastructure. Clouds have abstracted everything. They are a combination of IaaS and PaaS with additional features. They are controlled by APIs and softwares and can scale to a very large number of users. Apart from this, there were other benefits like:

• They can work with more complex technologies meant for automation, Internet of things, etc.
• The APIs could be managed by role based access making the clouds the most secure infrastructure so far.
• This is a scalable infrastructure in every sense. It provides scalability in terms of:
• Development
• Operation
• Provisioning and
• Maintenance

Cloud Native Infrastructure has reached us after a long chain of evolutions. During this course, the surrounding technologies have also evolved to be more flexible, efficient and technology agnostic. Native Infrastructure is necessary now more than ever. Working with distributed teams that are spread across multiple regions cannot be possible without a support like this. In the subsequent articles we will talk in detail about clouds, their core principles and their benefits. Stay tuned.