HomeChallengesThe Ownership Dilemma
learnbook poster
The Ownership DilemmaLast updated at Thu Oct 21 2021Skills
cloud-computing-basics
Tools
aws

Learning Goal


Understand how Cloud Computing can help organizations control Total Cost of Ownership for IT transformation projects


Let's read the below scenario posed by customer to understand the problem better.

The Problem Statement

Your client wants to have the discussion on TCO of the IT project. Based on that they would like to understand how Cloud can help them keep the TCO low. Can they reduce CAPEX? How about OPEX? Will cloud help them regulate that too?


How did we solve it?

To be able to suggest a viable answer, it is important to understand Total Cost of Ownership (TCO) first.

In the world of IT, the term TOC or Total Cost of Ownership is very commonly used term when it comes to estimating the full cost of implementing and maintaining a given digital transformation project.

Read more about TCO here and reflect a bit on your own experience


If you think a bit more about TCO, it appears similar to the scenarios below :

  • You are not an avid car enthusiast and only have limited travel needs. What should be your best decision in case you have a travel need?

    Buy a car straight away or take car on rent for your sporadic travel needs?

  • When you get pizza cravings would you go to the grocery store to purchase all raw materials, come back home and prepare the pizza painstakingly or simply ordering a pizza on a food delivery app?

The idea of Cloud computing is similar to the above 2 scenarios presented. Every cloud service provider is going to provide you a catalog of services available off the shelf which you can rent out and pay only for what you use.

This means the below :

  • Somebody else owns the machines/services
  • Somebody else takes the responsibility of maintenance
  • No need to hire expensive staff
  • Overall reduction in TCO. You own the process, not the infrastructure

Wow! Looks like this is exactly what you are looking for


So, we started wondering if cloud is that magic pill that would kill ou IT headache all together?


Well, as real as it gets, its not that straight forwarded.

There are several factors which come together to decide the level of dependency an organization would like to have on the Cloud providers.

  • Scenario 1 : A company having an app which generates humongous amount of data at high speeds would need to write their own distributed computing software and would only rent inter-connected, bare metal hardware hosted on a network of machines. Ex. Uber, Amazon, Twitter
  • Scenario 2 : A company having lot of regulations around security and sensitivity of data. Ex. Banks and Insurance companies. They simply can't put their databases on cloud straight away. They need to choose cautiously what applications they can host on cloud and what needs to stay on-premise
  • Scenario 3 :. New age startups who would happily go for cloud services which keeps their operating costs low. If that means hosting their entire application as well as data on remote servers, they are fine with it.

Considering the various kinds of dependency organizations would like to have on the cloud, leading service providers offer entire suite of services which could fit for most needs.

These offerings come under the labels of SaaS, PaaS, IaaS etc.

In order to best make sense of the above jargons, I would like you to have a look at the below image which explains which this brilliant analogy of pizza service. (Credits : Albert Barron, IBM)

Also, watch this easy to consume video to get a clear understanding of these terms :

(Credits : Bharti DW Consultancy)


Welcome back!

wave-hello-20.gif

Hope, you can now connect to the 3 scenarios I presented before.

With Cloud offering a number of services ranging from networking to storage to computing and many more, Traditional On-Premises set up is fast vanishing out of the IT landscape. It makes perfect sense to have as less CAPEX (Capital Expenditure) as possible to stay lean.

Moreover, depending upon the kind of services chosen by an organization, following would be the cost implication :

  • Organizations choosing for IaaS would end up paying the least to the cloud providers. All they are paying for is the hardware. Companies like Uber, Amazon would be a great fit for IaaS as they'll mostly need ready, connected and cheap commodity machines to build their solutions on

  • Organizations choosing for PaaS would end up paying for the usage of the platform for building solutions. This would be in addition to the infrastructure layer provided already as part of IaaS New age startups like Myntra, Dunzo etc. may also go for PaaS as they build working data and deployment pipelines using services directly from AWS, Azure or GCP.

  • Organizations choosing for SaaS pay up the most as their entire application alongwith data and infrastructure is hosted on cloud infrastructure. The startup themselves are Software offered as an app service. Ex, all commonly heard startup labels ex. Lyft, Uber, Byjus, Flipkart, Dunzo etc.


Now, coming back to our Problem Statement.

With the understanding of the kind of services offered by cloud, it would depend on what part of the application does the client wish to keep in-house (on-premise) and what part would they be ok moving to cloud. This would further depend on a number of important factors :

  • Criticality of Application - Organizations would like to keep their mission critical applications on Cloud since cloud service providers provide near 100% availability and scalability on demand
  • Sensitivity of Data - Organizations like Banks & Insurance due to regulatory compliances cannot move their data onto any third party machine. This scenario led to the birth of hybrid cloud set up where applications on cloud interact with data on-premises
  • Maturity of Business - Startups, in order to keep their barriers of exit low, would like to quickly set up, run, experiment and if failed, easily kill the systems. On the other hand, more matured organizations would like to keep a more hybrid approach to prevent themselves from vendor lock-in.

To Summarize :

  • Estimating TCO is important for organizations to understand how much the project is actually adding value to the business
  • The decision to move to cloud would depend on what components of the application is the customer ok with moving to a third party and how much it would need to own
  • In decreasing order of ownership, the kind of service models provided by the cloud service providers are : IaaS, PaaS, SaaS

Quiz Time!

Reflect a bit on your reading and take this quiz here to validate your learning

Wish to learn from a bigger community and work on an open project? Join us on Slack!


Created with 💙 by
author avatar
Amit Choudhary
Data Practitioner,Mentorskool
mentorskool logo
Mentee Today, Mentor Tomorrow
No 206, A Block, Sonesta Silver Oak,Varthur, Bangalore 560066,Karnataka
Copyright - 2022 © Mentorskool - All rights reserved.